{ "schema": "rapp-registry/1.0", "version": "1.0.0", "generated_at": "2026-11-12T01:14:14.851215+00:00", "stats": { "total_agents": 94, "publishers": 1, "categories": 13, "total_stacks": 73, "total_multi_agent_stacks": 2, "total_solution_containers": 71, "total_verticals": 13, "advertised_solutions": 72, "demo_proven_agents": 73, "source_verified_solutions": 72, "curated_copy_solutions": 72, "solution_onepagers": 72, "solution_demo_videos": 64, "total_lines": 51810, "total_kb": 2617.0, "total_first_party": 22, "first_party_available": 13, "first_party_preview": 9, "first_party_documented": 22, "first_party_products": [ "Dynamics 365 Business Central", "Dynamics 365 Contact Center", "Dynamics 365 Contact Center / Customer Service", "Dynamics 365 Customer Service", "Dynamics 365 Finance", "Dynamics 365 Project Operations", "Dynamics 365 Sales", "Dynamics 365 Supply Chain Management", "Microsoft 365 Copilot", "Microsoft Agent 365" ], "total_partner_agents": 12, "partner_list": [ "CongruentX" ], "partner_agent_categories": [ "b2b_sales", "financial_services", "professional_services", "retail_cpg" ], "publisher_list": [ "@aibast-agents-library" ], "category_list": [ "b2b_sales", "b2c_sales", "energy", "financial_services", "general", "healthcare", "human_resources", "manufacturing", "professional_services", "retail_cpg", "slg_government", "software_digital_products", "telecom_media_gaming" ], "tier_counts": { "verified": 92, "pilot": 2 } }, "stacks": [ { "stack": "account_intelligence", "vertical": "b2b_sales", "display_name": "Account Intelligence", "path": "agents/@aibast-agents-library/b2b_sales_stacks/account_intelligence_stack", "agents": [ "@aibast-agents-library/account-intelligence", "@aibast-agents-library/account-intelligence-orchestrator", "@aibast-agents-library/action-prioritization", "@aibast-agents-library/competitive-intelligence", "@aibast-agents-library/deal-tracking", "@aibast-agents-library/meeting-prep", "@aibast-agents-library/account-messaging", "@aibast-agents-library/account-risk-assessment", "@aibast-agents-library/stakeholder-intelligence" ], "agent_count": 9, "stack_type": "multi_agent" }, { "stack": "deal_progression", "vertical": "b2b_sales", "display_name": "Deal Progression", "path": "agents/@aibast-agents-library/b2b_sales_stacks/deal_progression_stack", "agents": [ "@aibast-agents-library/activity-gap", "@aibast-agents-library/deal-competitor-intel", "@aibast-agents-library/deal-health-score", "@aibast-agents-library/deal-progression", "@aibast-agents-library/deal-risk-assessment", "@aibast-agents-library/next-best-action", "@aibast-agents-library/pipeline-velocity", "@aibast-agents-library/revenue-forecast", "@aibast-agents-library/deal-stakeholder-engagement", "@aibast-agents-library/stalled-deal-detection", "@aibast-agents-library/win-probability" ], "agent_count": 11, "stack_type": "multi_agent" }, { "stack": "proposal_generation", "vertical": "b2b_sales", "display_name": "Proposal Generation", "path": "agents/@aibast-agents-library/b2b_sales_stacks/proposal_generation_stack", "agents": [ "@aibast-agents-library/proposal-generation" ], "agent_count": 1, "stack_type": "solution_container" }, { "stack": "sales_qualification", "vertical": "b2b_sales", "display_name": "Sales Qualification", "path": "agents/@aibast-agents-library/b2b_sales_stacks/sales_qualification_stack", "agents": [ "@aibast-agents-library/sales-qualification" ], "agent_count": 1, "stack_type": "solution_container" }, { "stack": "win_loss_analysis", "vertical": "b2b_sales", "display_name": "Win Loss Analysis", "path": "agents/@aibast-agents-library/b2b_sales_stacks/win_loss_analysis_stack", "agents": [ "@aibast-agents-library/win-loss-analysis" ], "agent_count": 1, "stack_type": "solution_container" }, { "stack": "cart_abandonment_recovery", "vertical": "b2c_sales", "display_name": "Cart Abandonment Recovery", "path": "agents/@aibast-agents-library/b2c_sales_stacks/cart_abandonment_recovery_stack", "agents": [ "@aibast-agents-library/cart-abandonment-recovery" ], "agent_count": 1, "stack_type": "solution_container" }, { "stack": "customer_loyalty_rewards", "vertical": "b2c_sales", "display_name": "Customer Loyalty Rewards", "path": "agents/@aibast-agents-library/b2c_sales_stacks/customer_loyalty_rewards_stack", "agents": [ "@aibast-agents-library/customer-loyalty-rewards" ], "agent_count": 1, "stack_type": "solution_container" }, { "stack": "omnichannel_engagement", "vertical": "b2c_sales", "display_name": "Omnichannel Engagement", "path": "agents/@aibast-agents-library/b2c_sales_stacks/omnichannel_engagement_stack", "agents": [ "@aibast-agents-library/omnichannel-engagement" ], "agent_count": 1, "stack_type": "solution_container" }, { "stack": "personalized_shopping_assistant", "vertical": "b2c_sales", "display_name": "Personalized Shopping Assistant", "path": "agents/@aibast-agents-library/b2c_sales_stacks/personalized_shopping_assistant_stack", "agents": [ "@aibast-agents-library/personalized-shopping-assistant" ], "agent_count": 1, "stack_type": "solution_container" }, { "stack": "asset_maintenance_forecast", "vertical": "energy", "display_name": "Asset Maintenance Forecast", "path": "agents/@aibast-agents-library/energy_stacks/asset_maintenance_forecast_stack", "agents": [ "@aibast-agents-library/asset-maintenance-forecast" ], "agent_count": 1, "stack_type": "solution_container" }, { "stack": "emission_tracking", "vertical": "energy", "display_name": "Emission Tracking", "path": "agents/@aibast-agents-library/energy_stacks/emission_tracking_stack", "agents": [ "@aibast-agents-library/emission-tracking" ], "agent_count": 1, "stack_type": "solution_container" }, { "stack": "engineering_standards_advisor", "vertical": "energy", "display_name": "Engineering Standards Advisor", "path": "agents/@aibast-agents-library/energy_stacks/engineering_standards_advisor_stack", "agents": [ "@aibast-agents-library/engineering-standards-advisor" ], "agent_count": 1, "stack_type": "solution_container" }, { "stack": "field_permit_to_work", "vertical": "energy", "display_name": "Field Permit To Work", "path": "agents/@aibast-agents-library/energy_stacks/field_permit_to_work_stack", "agents": [ "@aibast-agents-library/field-permit-to-work" ], "agent_count": 1, "stack_type": "solution_container" }, { "stack": "field_service_dispatch", "vertical": "energy", "display_name": "Field Service Dispatch", "path": "agents/@aibast-agents-library/energy_stacks/field_service_dispatch_stack", "agents": [ "@aibast-agents-library/field-service-dispatch" ], "agent_count": 1, "stack_type": "solution_container" }, { "stack": "grid_asset_maintenance", "vertical": "energy", "display_name": "Grid Asset Maintenance", "path": "agents/@aibast-agents-library/energy_stacks/grid_asset_maintenance_stack", "agents": [ "@aibast-agents-library/grid-asset-maintenance" ], "agent_count": 1, "stack_type": "solution_container" }, { "stack": "grid_outage_response", "vertical": "energy", "display_name": "Grid Outage Response", "path": "agents/@aibast-agents-library/energy_stacks/grid_outage_response_stack", "agents": [ "@aibast-agents-library/grid-outage-response" ], "agent_count": 1, "stack_type": "solution_container" }, { "stack": "permit_license_management", "vertical": "energy", "display_name": "Permit License Management", "path": "agents/@aibast-agents-library/energy_stacks/permit_license_management_stack", "agents": [ "@aibast-agents-library/permit-license-management" ], "agent_count": 1, "stack_type": "solution_container" }, { "stack": "procurement_supplier_collaboration", "vertical": "energy", "display_name": "Procurement Supplier Collaboration", "path": "agents/@aibast-agents-library/energy_stacks/procurement_supplier_collaboration_stack", "agents": [ "@aibast-agents-library/procurement-supplier-collaboration" ], "agent_count": 1, "stack_type": "solution_container" }, { "stack": "regulatory_reporting", "vertical": "energy", "display_name": "Regulatory Reporting", "path": "agents/@aibast-agents-library/energy_stacks/regulatory_reporting_stack", "agents": [ "@aibast-agents-library/energy-regulatory-reporting" ], "agent_count": 1, "stack_type": "solution_container" }, { "stack": "claims_processing", "vertical": "financial_services", "display_name": "Claims Processing", "path": "agents/@aibast-agents-library/financial_services_stacks/claims_processing_stack", "agents": [ "@aibast-agents-library/claims-processing" ], "agent_count": 1, "stack_type": "solution_container" }, { "stack": "customer_onboarding_fs", "vertical": "financial_services", "display_name": "Customer Onboarding Fs", "path": "agents/@aibast-agents-library/financial_services_stacks/customer_onboarding_fs_stack", "agents": [ "@aibast-agents-library/fs-customer-onboarding" ], "agent_count": 1, "stack_type": "solution_container" }, { "stack": "customer_sentiment_churn", "vertical": "financial_services", "display_name": "Customer Sentiment Churn", "path": "agents/@aibast-agents-library/financial_services_stacks/customer_sentiment_churn_stack", "agents": [ "@aibast-agents-library/customer-sentiment-churn" ], "agent_count": 1, "stack_type": "solution_container" }, { "stack": "financial_advisor_copilot", "vertical": "financial_services", "display_name": "Financial Advisor Copilot", "path": "agents/@aibast-agents-library/financial_services_stacks/financial_advisor_copilot_stack", "agents": [ "@aibast-agents-library/financial-advisor-copilot" ], "agent_count": 1, "stack_type": "solution_container" }, { "stack": "fraud_detection_alert", "vertical": "financial_services", "display_name": "Fraud Detection Alert", "path": "agents/@aibast-agents-library/financial_services_stacks/fraud_detection_alert_stack", "agents": [ "@aibast-agents-library/fraud-detection-alert" ], "agent_count": 1, "stack_type": "solution_container" }, { "stack": "loan_origination_assistant", "vertical": "financial_services", "display_name": "Loan Origination Assistant", "path": "agents/@aibast-agents-library/financial_services_stacks/loan_origination_assistant_stack", "agents": [ "@aibast-agents-library/loan-origination-assistant" ], "agent_count": 1, "stack_type": "solution_container" }, { "stack": "ma_target_sourcing", "vertical": "financial_services", "display_name": "Ma Target Sourcing", "path": "agents/@aibast-agents-library/financial_services_stacks/ma_target_sourcing_stack", "agents": [ "@aibast-agents-library/ma-target-sourcing" ], "agent_count": 1, "stack_type": "solution_container" }, { "stack": "payments_operations", "vertical": "financial_services", "display_name": "Payments Operations", "path": "agents/@aibast-agents-library/financial_services_stacks/payments_operations_stack", "agents": [ "@aibast-agents-library/payments-operations" ], "agent_count": 1, "stack_type": "solution_container" }, { "stack": "portfolio_rebalancing", "vertical": "financial_services", "display_name": "Portfolio Rebalancing", "path": "agents/@aibast-agents-library/financial_services_stacks/portfolio_rebalancing_stack", "agents": [ "@aibast-agents-library/portfolio-rebalancing" ], "agent_count": 1, "stack_type": "solution_container" }, { "stack": "regulatory_compliance_fs", "vertical": "financial_services", "display_name": "Regulatory Compliance Fs", "path": "agents/@aibast-agents-library/financial_services_stacks/regulatory_compliance_fs_stack", "agents": [ "@aibast-agents-library/fs-regulatory-compliance" ], "agent_count": 1, "stack_type": "solution_container" }, { "stack": "reporting_package_validation", "vertical": "financial_services", "display_name": "Reporting Package Validation", "path": "agents/@aibast-agents-library/financial_services_stacks/reporting_package_validation_stack", "agents": [ "@aibast-agents-library/reporting-package-validation" ], "agent_count": 1, "stack_type": "solution_container" }, { "stack": "underwriting_support", "vertical": "financial_services", "display_name": "Underwriting Support", "path": "agents/@aibast-agents-library/financial_services_stacks/underwriting_support_stack", "agents": [ "@aibast-agents-library/underwriting-support" ], "agent_count": 1, "stack_type": "solution_container" }, { "stack": "wealth_insights_generator", "vertical": "financial_services", "display_name": "Wealth Insights Generator", "path": "agents/@aibast-agents-library/financial_services_stacks/wealth_insights_generator_stack", "agents": [ "@aibast-agents-library/wealth-insights-generator" ], "agent_count": 1, "stack_type": "solution_container" }, { "stack": "ai_customer_assistant", "vertical": "general", "display_name": "Ai Customer Assistant", "path": "agents/@aibast-agents-library/general_stacks/ai_customer_assistant_stack", "agents": [ "@aibast-agents-library/ai-customer-assistant" ], "agent_count": 1, "stack_type": "solution_container" }, { "stack": "cross_selling_opportunities", "vertical": "general", "display_name": "Cross Selling Opportunities", "path": "agents/@aibast-agents-library/general_stacks/cross_selling_opportunities_stack", "agents": [ "@aibast-agents-library/cross-selling" ], "agent_count": 1, "stack_type": "solution_container" }, { "stack": "procurement_agent", "vertical": "general", "display_name": "Procurement Agent", "path": "agents/@aibast-agents-library/general_stacks/procurement_agent_stack", "agents": [ "@aibast-agents-library/procurement-agent" ], "agent_count": 1, "stack_type": "solution_container" }, { "stack": "procurement_support", "vertical": "general", "display_name": "Procurement Support", "path": "agents/@aibast-agents-library/general_stacks/procurement_support_stack", "agents": [ "@aibast-agents-library/procurement-support" ], "agent_count": 1, "stack_type": "solution_container" }, { "stack": "care_gap_closure", "vertical": "healthcare", "display_name": "Care Gap Closure", "path": "agents/@aibast-agents-library/healthcare_stacks/care_gap_closure_stack", "agents": [ "@aibast-agents-library/care-gap-closure" ], "agent_count": 1, "stack_type": "solution_container" }, { "stack": "clinical_notes_summarizer", "vertical": "healthcare", "display_name": "Clinical Notes Summarizer", "path": "agents/@aibast-agents-library/healthcare_stacks/clinical_notes_summarizer_stack", "agents": [ "@aibast-agents-library/clinical-notes-summarizer" ], "agent_count": 1, "stack_type": "solution_container" }, { "stack": "patient_intake", "vertical": "healthcare", "display_name": "Patient Intake", "path": "agents/@aibast-agents-library/healthcare_stacks/patient_intake_stack", "agents": [ "@aibast-agents-library/patient-intake" ], "agent_count": 1, "stack_type": "solution_container" }, { "stack": "prior_authorization", "vertical": "healthcare", "display_name": "Prior Authorization", "path": "agents/@aibast-agents-library/healthcare_stacks/prior_authorization_stack", "agents": [ "@aibast-agents-library/prior-authorization" ], "agent_count": 1, "stack_type": "solution_container" }, { "stack": "tiered_huddle_board", "vertical": "healthcare", "display_name": "Tiered Huddle Board", "path": "agents/@aibast-agents-library/healthcare_stacks/tiered_huddle_board_stack", "agents": [ "@aibast-agents-library/tiered-huddle-board" ], "agent_count": 1, "stack_type": "solution_container" }, { "stack": "ask_hr", "vertical": "human_resources", "display_name": "Ask Hr", "path": "agents/@aibast-agents-library/human_resources_stacks/ask_hr_stack", "agents": [ "@aibast-agents-library/ask-hr" ], "agent_count": 1, "stack_type": "solution_container" }, { "stack": "benefits_enrollment_assistant", "vertical": "human_resources", "display_name": "Benefits Enrollment Assistant", "path": "agents/@aibast-agents-library/human_resources_stacks/benefits_enrollment_assistant_stack", "agents": [ "@aibast-agents-library/benefits-enrollment-assistant" ], "agent_count": 1, "stack_type": "solution_container" }, { "stack": "build_materials_compliance", "vertical": "manufacturing", "display_name": "Build Materials Compliance", "path": "agents/@aibast-agents-library/manufacturing_stacks/build_materials_compliance_stack", "agents": [ "@aibast-agents-library/build-materials-compliance" ], "agent_count": 1, "stack_type": "solution_container" }, { "stack": "inventory_rebalancing", "vertical": "manufacturing", "display_name": "Inventory Rebalancing", "path": "agents/@aibast-agents-library/manufacturing_stacks/inventory_rebalancing_stack", "agents": [ "@aibast-agents-library/inventory-rebalancing" ], "agent_count": 1, "stack_type": "solution_container" }, { "stack": "maintenance_scheduling", "vertical": "manufacturing", "display_name": "Maintenance Scheduling", "path": "agents/@aibast-agents-library/manufacturing_stacks/maintenance_scheduling_stack", "agents": [ "@aibast-agents-library/maintenance-scheduling" ], "agent_count": 1, "stack_type": "solution_container" }, { "stack": "order_entry_validation", "vertical": "manufacturing", "display_name": "Order Entry Validation", "path": "agents/@aibast-agents-library/manufacturing_stacks/order_entry_validation_stack", "agents": [ "@aibast-agents-library/order-entry-validation" ], "agent_count": 1, "stack_type": "solution_container" }, { "stack": "order_status_communication", "vertical": "manufacturing", "display_name": "Order Status Communication", "path": "agents/@aibast-agents-library/manufacturing_stacks/order_status_communication_stack", "agents": [ "@aibast-agents-library/order-status-communication" ], "agent_count": 1, "stack_type": "solution_container" }, { "stack": "production_line_optimization", "vertical": "manufacturing", "display_name": "Production Line Optimization", "path": "agents/@aibast-agents-library/manufacturing_stacks/production_line_optimization_stack", "agents": [ "@aibast-agents-library/production-line-optimization" ], "agent_count": 1, "stack_type": "solution_container" }, { "stack": "supplier_risk_monitoring", "vertical": "manufacturing", "display_name": "Supplier Risk Monitoring", "path": "agents/@aibast-agents-library/manufacturing_stacks/supplier_risk_monitoring_stack", "agents": [ "@aibast-agents-library/supplier-risk-monitoring" ], "agent_count": 1, "stack_type": "solution_container" }, { "stack": "warranty_registration", "vertical": "manufacturing", "display_name": "Warranty Registration", "path": "agents/@aibast-agents-library/manufacturing_stacks/warranty_registration_stack", "agents": [ "@aibast-agents-library/warranty-registration" ], "agent_count": 1, "stack_type": "solution_container" }, { "stack": "client_health_score", "vertical": "professional_services", "display_name": "Client Health Score", "path": "agents/@aibast-agents-library/professional_services_stacks/client_health_score_stack", "agents": [ "@aibast-agents-library/client-health-score" ], "agent_count": 1, "stack_type": "solution_container" }, { "stack": "contract_risk_review", "vertical": "professional_services", "display_name": "Contract Risk Review", "path": "agents/@aibast-agents-library/professional_services_stacks/contract_risk_review_stack", "agents": [ "@aibast-agents-library/contract-risk-review" ], "agent_count": 1, "stack_type": "solution_container" }, { "stack": "resource_utilization", "vertical": "professional_services", "display_name": "Resource Utilization", "path": "agents/@aibast-agents-library/professional_services_stacks/resource_utilization_stack", "agents": [ "@aibast-agents-library/resource-utilization" ], "agent_count": 1, "stack_type": "solution_container" }, { "stack": "solution_design_review", "vertical": "professional_services", "display_name": "Solution Design Review", "path": "agents/@aibast-agents-library/professional_services_stacks/solution_design_review_stack", "agents": [ "@aibast-agents-library/solution-design-review" ], "agent_count": 1, "stack_type": "solution_container" }, { "stack": "staffing_territory_planning", "vertical": "professional_services", "display_name": "Staffing Territory Planning", "path": "agents/@aibast-agents-library/professional_services_stacks/staffing_territory_planning_stack", "agents": [ "@aibast-agents-library/staffing-territory-planning" ], "agent_count": 1, "stack_type": "solution_container" }, { "stack": "time_entry_billing", "vertical": "professional_services", "display_name": "Time Entry Billing", "path": "agents/@aibast-agents-library/professional_services_stacks/time_entry_billing_stack", "agents": [ "@aibast-agents-library/time-entry-billing" ], "agent_count": 1, "stack_type": "solution_container" }, { "stack": "inventory_visibility", "vertical": "retail_cpg", "display_name": "Inventory Visibility", "path": "agents/@aibast-agents-library/retail_cpg_stacks/inventory_visibility_stack", "agents": [ "@aibast-agents-library/inventory-visibility" ], "agent_count": 1, "stack_type": "solution_container" }, { "stack": "personalized_marketing", "vertical": "retail_cpg", "display_name": "Personalized Marketing", "path": "agents/@aibast-agents-library/retail_cpg_stacks/personalized_marketing_stack", "agents": [ "@aibast-agents-library/personalized-marketing" ], "agent_count": 1, "stack_type": "solution_container" }, { "stack": "returns_complaints_resolution", "vertical": "retail_cpg", "display_name": "Returns Complaints Resolution", "path": "agents/@aibast-agents-library/retail_cpg_stacks/returns_complaints_resolution_stack", "agents": [ "@aibast-agents-library/returns-complaints-resolution" ], "agent_count": 1, "stack_type": "solution_container" }, { "stack": "store_associate_copilot", "vertical": "retail_cpg", "display_name": "Store Associate Copilot", "path": "agents/@aibast-agents-library/retail_cpg_stacks/store_associate_copilot_stack", "agents": [ "@aibast-agents-library/store-associate-copilot" ], "agent_count": 1, "stack_type": "solution_container" }, { "stack": "supplier_claim_engagement", "vertical": "retail_cpg", "display_name": "Supplier Claim Engagement", "path": "agents/@aibast-agents-library/retail_cpg_stacks/supplier_claim_engagement_stack", "agents": [ "@aibast-agents-library/supplier-claim-engagement" ], "agent_count": 1, "stack_type": "solution_container" }, { "stack": "supply_chain_disruption_alert", "vertical": "retail_cpg", "display_name": "Supply Chain Disruption Alert", "path": "agents/@aibast-agents-library/retail_cpg_stacks/supply_chain_disruption_alert_stack", "agents": [ "@aibast-agents-library/supply-chain-disruption-alert" ], "agent_count": 1, "stack_type": "solution_container" }, { "stack": "briefing_pack_builder", "vertical": "slg_government", "display_name": "Briefing Pack Builder", "path": "agents/@aibast-agents-library/slg_government_stacks/briefing_pack_builder_stack", "agents": [ "@aibast-agents-library/briefing-pack-builder" ], "agent_count": 1, "stack_type": "solution_container" }, { "stack": "building_permit_processing", "vertical": "slg_government", "display_name": "Building Permit Processing", "path": "agents/@aibast-agents-library/slg_government_stacks/building_permit_processing_stack", "agents": [ "@aibast-agents-library/building-permit-processing" ], "agent_count": 1, "stack_type": "solution_container" }, { "stack": "public_correspondence_response", "vertical": "slg_government", "display_name": "Public Correspondence Response", "path": "agents/@aibast-agents-library/slg_government_stacks/public_correspondence_response_stack", "agents": [ "@aibast-agents-library/public-correspondence-response" ], "agent_count": 1, "stack_type": "solution_container" }, { "stack": "utility_billing_assistance", "vertical": "slg_government", "display_name": "Utility Billing Assistance", "path": "agents/@aibast-agents-library/slg_government_stacks/utility_billing_assistance_stack", "agents": [ "@aibast-agents-library/utility-billing-assistance" ], "agent_count": 1, "stack_type": "solution_container" }, { "stack": "license_renewal_expansion", "vertical": "software_digital_products", "display_name": "License Renewal Expansion", "path": "agents/@aibast-agents-library/software_dp_stacks/license_renewal_expansion_stack", "agents": [ "@aibast-agents-library/license-renewal-expansion" ], "agent_count": 1, "stack_type": "solution_container" }, { "stack": "product_feedback_synthesizer", "vertical": "software_digital_products", "display_name": "Product Feedback Synthesizer", "path": "agents/@aibast-agents-library/software_dp_stacks/product_feedback_synthesizer_stack", "agents": [ "@aibast-agents-library/product-feedback-synthesizer" ], "agent_count": 1, "stack_type": "solution_container" }, { "stack": "campaign_delivery_optimizer", "vertical": "telecom_media_gaming", "display_name": "Campaign Delivery Optimizer", "path": "agents/@aibast-agents-library/telecom_media_gaming_stacks/campaign_delivery_optimizer_stack", "agents": [ "@aibast-agents-library/campaign-delivery-optimizer" ], "agent_count": 1, "stack_type": "solution_container" }, { "stack": "content_launch_clearance", "vertical": "telecom_media_gaming", "display_name": "Content Launch Clearance", "path": "agents/@aibast-agents-library/telecom_media_gaming_stacks/content_launch_clearance_stack", "agents": [ "@aibast-agents-library/content-launch-clearance" ], "agent_count": 1, "stack_type": "solution_container" }, { "stack": "first_time_fix", "vertical": "telecom_media_gaming", "display_name": "First Time Fix", "path": "agents/@aibast-agents-library/telecom_media_gaming_stacks/first_time_fix_stack", "agents": [ "@aibast-agents-library/first-time-fix" ], "agent_count": 1, "stack_type": "solution_container" }, { "stack": "live_event_launch_planner", "vertical": "telecom_media_gaming", "display_name": "Live Event Launch Planner", "path": "agents/@aibast-agents-library/telecom_media_gaming_stacks/live_event_launch_planner_stack", "agents": [ "@aibast-agents-library/live-event-launch-planner" ], "agent_count": 1, "stack_type": "solution_container" } ], "agents": [ { "schema": "rapp-agent/1.0", "name": "@aibast-agents-library/account-intelligence", "version": "1.0.0", "display_name": "Account Intelligence Agent", "description": "Automate account research and strategy planning to help sellers prepare faster, win more, and elevate deal quality.", "author": "AIBAST", "tags": [ "b2b", "sales", "account-intelligence", "stakeholder-mapping", "competitive-intel" ], "category": "b2b_sales", "quality_tier": "verified", "requires_env": [], "dependencies": [ "@rapp/basic-agent" ], "_file": "agents/@aibast-agents-library/b2b_sales_stacks/account_intelligence_stack/account_intelligence_agent.py", "_size_kb": 28.6, "_lines": 491, "_sha256": "1a8149d27ccfaf4bfc8ab892afcada0378d9c0ff2fd514f0f53a4c39d93c1f87", "_install_prefix": "account_intelligence__", "_install_filename": "account_intelligence__1a8149d27ccf_agent.py", "_stack": "account_intelligence", "_stack_vertical": "b2b_sales", "_synthetic_data": true, "_added_at": "2026-03-18T13:23:12-04:00", "_solution": { "advertised_name": "Account Intelligence Agent", "sharepoint_list_name": "Account Research Agent", "slot": 19, "executive_summary": "Automate account research and strategy planning to help sellers prepare faster, win more, and elevate deal quality.", "industries": [ "Cross-Industry" ], "personas": [ "Account Executive", "Sales Director", "Customer Success Manager" ], "featured_tools": [ "Dynamics 365", "SharePoint", "Microsoft Teams" ], "agent_requirements": [ "Microsoft Copilot Studio" ], "capabilities": [ "Get a single unified view of account signals across diverse sources", "Create stakeholder maps with influence scores/relationship gaps", "Deliver competitive insights and risk mitigation strategies", "Produced talking points and pre-meeting checklists" ], "outcomes": [ "Improved win rates via better stakeholder coverage and competitive positioning", "Reduced account research time from hours to minutes", "Elevated deal quality with tailored messaging and risk mitigation", "Enabled real-time team alignment with automated alerts and updates" ], "customer_scenario": [ "Account research demanded hours of manual research across sources", "Incomplete stakeholder mapping missed key decision makers", "Proposals lacked full differentiation without full competitive context", "Sellers lacked tailored, persona-based messaging and resources" ], "slide_title": "Account Intelligence Agent", "slide_summary": "Automate account research and strategy planning to help sellers prepare faster, win more, and elevate deal quality.", "scenario_name": "A global technology company", "challenge_intro": "An enterprise sales team struggled to prepare effectively for meetings:", "actions_intro": "The technology company used the Account Intelligence Agent to:", "outcomes_intro": "With the Agent, the company:", "opportunity_statements": [ "Aggregate business intelligence to uncover sales opportunities in minutes", "Perform stakeholder mapping for visibility into purchasing paths and engagement gaps", "Deliver competitive analysis and positioning recommendations to help sellers effectively differentiate", "Create personalized engagement strategies to empower sellers for every conversation" ], "module_name": null, "module_description": null, "source_verified": true, "sales_headline": "Prepare for the account conversation from one evidence view", "card_pitch": "Bring account context, stakeholder coverage, competitor signals, risks, and draft talking points into a single reviewable briefing for sellers and customer-success teams.", "why_try": "A seller can ask what matters before a meeting and receive both the evidence and the gaps that need human follow-up.", "customer_challenge_copy": "Account preparation is spread across records, relationship history, news, competitor notes, and seller memory, making coverage inconsistent.", "microsoft_ai_story": "Microsoft Copilot Studio provides a natural-language account briefing. Dynamics 365 Sales contributes account and relationship context, SharePoint supplies account plans and approved references, and Microsoft Teams supports seller collaboration and meeting preparation.", "business_value": [ "Creates a more consistent account-research and meeting-preparation workflow.", "Makes stakeholder and relationship gaps easier to review.", "Improves positioning and risk discussions without automating engagement." ], "search_terms": [ "account intelligence", "account research", "stakeholder mapping", "buying committee", "competitive intelligence", "meeting preparation", "deal risk", "Dynamics 365 Sales", "SharePoint account plans", "Microsoft Teams sellers", "customer success", "Copilot Studio" ], "journey_stage": "Understand", "blueprint_role": "Creates a read-only account intelligence layer for meeting preparation, relationship planning, and competitive review.", "sample_prompts": [ { "label": "Review the account", "prompt": "Prepare a synthetic Acme Corporation account overview and show the evidence that deserves seller attention.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=account-intelligence-account-overview" }, { "label": "Map the buying committee", "prompt": "Map the synthetic Acme buying committee and show the relationship gaps without creating outreach tasks.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=account-intelligence-stakeholder-map" }, { "label": "Compare competitor signals", "prompt": "Compare the synthetic competitor signals and positioning considerations for Acme Corporation.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=account-intelligence-competitive-intel" }, { "label": "Draft meeting talking points", "prompt": "Draft persona-specific talking points for the synthetic Acme meeting; do not send messages.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=account-intelligence-value-messaging" }, { "label": "Assess deal risks", "prompt": "Assess the synthetic Acme deal risks and mitigation options without changing a forecast or CRM record.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=account-intelligence-risk-assessment" }, { "label": "Prepare the executive briefing", "prompt": "Prepare a concise synthetic Acme pre-meeting briefing and review checklist.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=account-intelligence-executive-briefing" } ], "architecture": { "business_flow": [ "Account executive, sales director, or customer success manager", "Copilot experience", "Account Intelligence Agent", "Synthetic account evidence", "Human meeting and account planning" ], "capabilities": [ { "operation": "account_overview", "name": "Account Overview", "purpose": "Combines firmographic, adoption, engagement, support, and activity evidence." }, { "operation": "stakeholder_map", "name": "Stakeholder Map", "purpose": "Surfaces buying roles, influence, sentiment, engagement, and relationship gaps." }, { "operation": "competitive_intel", "name": "Competitive Intel", "purpose": "Compares synthetic competitor activity and positioning considerations." }, { "operation": "value_messaging", "name": "Value Messaging", "purpose": "Drafts persona-aligned talking points for human review." }, { "operation": "risk_assessment", "name": "Risk Assessment", "purpose": "Explains deal risks and candidate mitigations without changing a forecast." }, { "operation": "executive_briefing", "name": "Executive Briefing", "purpose": "Compiles a concise pre-meeting view and checklist." } ], "easy_mode": [ "Verify Power Platform CLI and authenticate to the target environment.", "Initialize a modern cli-copilot project with an approved publisher prefix.", "Upload the two synthetic knowledge files and create one manual skill for every advertised operation.", "Bind dedicated approved connection references only after replacing the synthetic evidence route.", "Push the draft, replay every locked persona case, and stop before publication or any external action." ], "local_install_prompt": "You are GitHub Copilot running in Agent mode in VS Code. Install and open the AI BAST Account Intelligence Agent from this deployment recipe: https://raw.githubusercontent.com/microsoft/aibast-agents-library/main/solutions/account-intelligence/deployment.json. Own the setup and do not ask me to run terminal commands. Inspect the operating system; verify the Brainstem health endpoint; if needed, use the official installer from the recipe; fetch and install the portable source; verify the expected tool; and replay the recipe smoke test. Preserve the synthetic-only route for account_overview, stakeholder_map, competitive_intel, value_messaging, risk_assessment, executive_briefing. Do not connect customer systems, perform writes, publish, approve pricing, send outreach, or silently substitute another agent. If authentication requires browser confirmation, ask only for that browser step. Stop and report the exact evidence and any failed step. Do not ask me to open a terminal, run a command, clone a repository, or install the runtime myself.", "copilot_studio_prompt": "You are GitHub Copilot running in Agent mode in VS Code with the Microsoft Copilot Studio plugin installed. Prepare the AI BAST Account Intelligence Agent from https://raw.githubusercontent.com/microsoft/aibast-agents-library/main/solutions/account-intelligence/deployment.json. Ask only for the target Power Platform environment ID and publisher prefix. Verify PAC CLI, authenticate if needed, initialize a cli-copilot project, upload the two synthetic knowledge files, and create one skill for each operation: account_overview, stakeholder_map, competitive_intel, value_messaging, risk_assessment, executive_briefing. Bind only dedicated approved connection references, preserve every evidence and approval boundary, pull before push, and replay every locked persona case. Stop before publish and show the generated files, validation evidence, unresolved connection work, and the exact publish command that would require user approval.", "required_connections": [ "Dynamics 365 Sales", "SharePoint", "Microsoft Teams", "Approved business-intelligence source" ], "manual_commands": [ "pac auth create", "pac copilot init --name \"Account Intelligence Agent\" --publisher-prefix --authoring-mode cli-copilot --project-dir \"\" --environment \"\"", "pac connection list --environment \"\"", "pac copilot pull --project-dir \"\"", "pac copilot push --project-dir \"\"", "pac copilot publish --bot \"\" --environment \"\"" ], "acceptance_checks": [ "The package remains under solutions/account-intelligence and the portable source resolves to the expected tool.", "Every advertised operation has one uploadable skill and one locked persona-language capture case.", "The approved one-pager filename and source hash remain pinned in the one-pager map.", "Every exact value is identified as synthetic or as a planning assumption rather than a measured customer outcome.", "No CRM write, outreach, approval, pricing change, forecast change, task, alert, document delivery, or customer communication occurs.", "Publishing remains a separate user-approved step after push and validation." ], "hard_mode": [ "Run pac copilot init and confirm settings.mcs.yml preserves the intended environment, language, and publisher identity.", "Author global instructions that route only the supported operations: account_overview, stakeholder_map, competitive_intel, value_messaging, risk_assessment, executive_briefing.", "Upload both synthetic knowledge files and create each operation skill from the package SKILL.md source.", "Create dedicated connection references for the approved production seams; do not reuse ambiguous existing connections.", "Validate schema, instructions, skills, knowledge, and connection bindings; pull before push and replay every locked case.", "Keep the agent in Draft and require explicit confirmation before publish or any external action." ] }, "curated_copy": true, "source": "catalog", "origin": null, "onepager": "Account Intelligence Agent one-pager.pptx", "demo_video": "10-Account Intelligence Agent.mp4", "has_onepager": true, "has_demo_video": true, "promise_coverage": 80, "is_primary": true, "aliases": [], "package": { "slug": "account-intelligence", "quest_url": "solutions/account-intelligence/quest.html", "manual_tutorial_url": "solutions/account-intelligence/manual-tutorial.html", "export_manifest_url": "solutions/account-intelligence/export-manifest.json" } }, "_demo": { "slug": "account-intelligence", "case_count": 6, "personas": [ "Account Executive", "Customer Success Manager", "Sales Director" ], "prompts": [ "Prepare a synthetic Acme Corporation account overview and show the evidence that deserves seller attention.", "Map the synthetic Acme buying committee and show the relationship gaps without creating outreach tasks.", "Compare the synthetic competitor signals and positioning considerations for Acme Corporation." ] }, "_catalog_kind": "solution", "_readiness": [ "Demo proven", "Conversation tested", "Synthetic data", "Production pattern" ] }, { "schema": "rapp-agent/1.0", "name": "@aibast-agents-library/account-intelligence-orchestrator", "version": "1.0.0", "display_name": "Account Intelligence Orchestrator", "description": "Coordinates multi-agent pipelines for 360-degree account intelligence briefings.", "author": "AIBAST", "tags": [ "b2b", "sales", "orchestration", "pipeline", "account-intelligence" ], "category": "b2b_sales", "quality_tier": "verified", "requires_env": [], "dependencies": [ "@rapp/basic-agent" ], "_file": "agents/@aibast-agents-library/b2b_sales_stacks/account_intelligence_stack/account_intelligence_orchestrator.py", "_size_kb": 20.8, "_lines": 457, "_sha256": "a21f50d0b417a4cd1f78d6e5cd4a3a19bf946ea1531f05708c28a77f5e789aab", "_install_prefix": "account_intelligence_orchestrator__", "_install_filename": "account_intelligence_orchestrator__a21f50d0b417_agent.py", "_stack": "account_intelligence", "_stack_vertical": "b2b_sales", "_synthetic_data": true, "_added_at": "2026-03-18T13:23:12-04:00", "_catalog_kind": "component", "_readiness": [ "Synthetic data" ] }, { "schema": "rapp-agent/1.0", "name": "@aibast-agents-library/action-prioritization", "version": "1.0.0", "display_name": "Action Prioritization", "description": "Prioritizes sales actions, generates daily plans, and optimizes resource allocation across deals.", "author": "AIBAST", "tags": [ "b2b", "sales", "action-prioritization", "planning", "resource-allocation" ], "category": "b2b_sales", "quality_tier": "verified", "requires_env": [], "dependencies": [ "@rapp/basic-agent" ], "_file": "agents/@aibast-agents-library/b2b_sales_stacks/account_intelligence_stack/action_prioritization_agent.py", "_size_kb": 19.7, "_lines": 398, "_sha256": "9e6deed0078a65dfa4f73a7a7bde65080d9d97b513efb99129400750ed3f6030", "_install_prefix": "action_prioritization__", "_install_filename": "action_prioritization__9e6deed0078a_agent.py", "_stack": "account_intelligence", "_stack_vertical": "b2b_sales", "_synthetic_data": true, "_added_at": "2026-03-18T13:23:12-04:00", "_catalog_kind": "component", "_readiness": [ "Synthetic data" ] }, { "schema": "rapp-agent/1.0", "name": "@aibast-agents-library/competitive-intelligence", "version": "1.0.0", "display_name": "Competitive Intelligence", "description": "Analyzes competitive landscapes, win/loss patterns, and generates battlecards for enterprise deals.", "author": "AIBAST", "tags": [ "b2b", "sales", "competitive-intelligence", "battlecards", "win-loss" ], "category": "b2b_sales", "quality_tier": "verified", "requires_env": [], "dependencies": [ "@rapp/basic-agent" ], "_file": "agents/@aibast-agents-library/b2b_sales_stacks/account_intelligence_stack/competitive_intelligence_agent.py", "_size_kb": 21.3, "_lines": 441, "_sha256": "8848b9a389d6a017a30b98dbd62a2713e6086b41e6a25b7b2b91e77ff0420556", "_install_prefix": "competitive_intelligence__", "_install_filename": "competitive_intelligence__8848b9a389d6_agent.py", "_stack": "account_intelligence", "_stack_vertical": "b2b_sales", "_synthetic_data": true, "_added_at": "2026-03-18T13:23:12-04:00", "_catalog_kind": "component", "_readiness": [ "Synthetic data" ] }, { "schema": "rapp-agent/1.0", "name": "@aibast-agents-library/deal-tracking", "version": "1.0.0", "display_name": "Deal Tracking", "description": "Tracks pipeline, deal movement, stage velocity, and forecast accuracy for enterprise sales.", "author": "AIBAST", "tags": [ "b2b", "sales", "deal-tracking", "pipeline", "forecasting" ], "category": "b2b_sales", "quality_tier": "verified", "requires_env": [], "dependencies": [ "@rapp/basic-agent" ], "_file": "agents/@aibast-agents-library/b2b_sales_stacks/account_intelligence_stack/deal_tracking_agent.py", "_size_kb": 18.6, "_lines": 402, "_sha256": "19d403a2a7d46a308800c3a54cf62d19eddd104242673f00bd0051d8665ff4c8", "_install_prefix": "deal_tracking__", "_install_filename": "deal_tracking__19d403a2a7d4_agent.py", "_stack": "account_intelligence", "_stack_vertical": "b2b_sales", "_synthetic_data": true, "_added_at": "2026-03-18T13:23:12-04:00", "_catalog_kind": "component", "_readiness": [ "Synthetic data" ] }, { "schema": "rapp-agent/1.0", "name": "@aibast-agents-library/meeting-prep", "version": "1.0.0", "display_name": "Meeting Prep", "description": "Generates pre-meeting briefs, talking points, objection prep, and follow-up templates.", "author": "AIBAST", "tags": [ "b2b", "sales", "meeting-prep", "briefing", "objection-handling" ], "category": "b2b_sales", "quality_tier": "verified", "requires_env": [], "dependencies": [ "@rapp/basic-agent" ], "_file": "agents/@aibast-agents-library/b2b_sales_stacks/account_intelligence_stack/meeting_prep_agent.py", "_size_kb": 25.1, "_lines": 521, "_sha256": "0b0d029fe2dc9b8de15bd906905d5fc23e3dba72b264630b3cc7d5a0586c7275", "_install_prefix": "meeting_prep__", "_install_filename": "meeting_prep__0b0d029fe2dc_agent.py", "_stack": "account_intelligence", "_stack_vertical": "b2b_sales", "_synthetic_data": true, "_added_at": "2026-03-18T13:23:12-04:00", "_catalog_kind": "component", "_readiness": [ "Synthetic data" ] }, { "schema": "rapp-agent/1.0", "name": "@aibast-agents-library/account-messaging", "version": "1.0.0", "display_name": "Account Messaging", "description": "Generates personalized outreach, follow-ups, and campaign sequences for enterprise accounts.", "author": "AIBAST", "tags": [ "b2b", "sales", "messaging", "outreach", "email-sequences" ], "category": "b2b_sales", "quality_tier": "verified", "requires_env": [], "dependencies": [ "@rapp/basic-agent" ], "_file": "agents/@aibast-agents-library/b2b_sales_stacks/account_intelligence_stack/messaging_agent.py", "_size_kb": 20.3, "_lines": 429, "_sha256": "f0fdb371c236ea54cab996b385a744fcd20395459f2b5d5b313721b87ca9eb35", "_install_prefix": "account_messaging__", "_install_filename": "account_messaging__f0fdb371c236_agent.py", "_stack": "account_intelligence", "_stack_vertical": "b2b_sales", "_synthetic_data": true, "_added_at": "2026-03-18T13:23:12-04:00", "_catalog_kind": "component", "_readiness": [ "Synthetic data" ] }, { "schema": "rapp-agent/1.0", "name": "@aibast-agents-library/account-risk-assessment", "version": "1.0.0", "display_name": "Account Risk Assessment", "description": "Assesses deal risk, churn probability, and financial health with mitigation recommendations.", "author": "AIBAST", "tags": [ "b2b", "sales", "risk-assessment", "churn-prediction", "deal-risk" ], "category": "b2b_sales", "quality_tier": "verified", "requires_env": [], "dependencies": [ "@rapp/basic-agent" ], "_file": "agents/@aibast-agents-library/b2b_sales_stacks/account_intelligence_stack/risk_assessment_agent.py", "_size_kb": 23.0, "_lines": 495, "_sha256": "ea496765dbd7aad657f3531af2da1c05cefe994ddbec9e5d8307e3d255c4bb0c", "_install_prefix": "account_risk_assessment__", "_install_filename": "account_risk_assessment__ea496765dbd7_agent.py", "_stack": "account_intelligence", "_stack_vertical": "b2b_sales", "_synthetic_data": true, "_added_at": "2026-03-18T13:23:12-04:00", "_catalog_kind": "component", "_readiness": [ "Synthetic data" ] }, { "schema": "rapp-agent/1.0", "name": "@aibast-agents-library/stakeholder-intelligence", "version": "1.0.0", "display_name": "Stakeholder Intelligence", "description": "Maps org charts, analyzes buying committees, and scores relationship strength for enterprise accounts.", "author": "AIBAST", "tags": [ "b2b", "sales", "stakeholder-mapping", "org-chart", "buying-committee" ], "category": "b2b_sales", "quality_tier": "verified", "requires_env": [], "dependencies": [ "@rapp/basic-agent" ], "_file": "agents/@aibast-agents-library/b2b_sales_stacks/account_intelligence_stack/stakeholder_intelligence_agent.py", "_size_kb": 23.4, "_lines": 440, "_sha256": "1100b908472d7ed850f5263503d682f686ba0af05f5b5b1f951c76f143cb05c0", "_install_prefix": "stakeholder_intelligence__", "_install_filename": "stakeholder_intelligence__1100b908472d_agent.py", "_stack": "account_intelligence", "_stack_vertical": "b2b_sales", "_synthetic_data": true, "_added_at": "2026-03-18T13:23:12-04:00", "_catalog_kind": "component", "_readiness": [ "Synthetic data" ] }, { "schema": "rapp-agent/1.0", "name": "@aibast-agents-library/activity-gap", "version": "1.0.0", "display_name": "Activity Gap Analyzer", "description": "Identifies missing sales activities per stage, evaluates completion, generates roadmaps, and analyzes gap impact.", "author": "AIBAST", "tags": [ "b2b", "sales", "activity-gap", "deal-progression", "pipeline" ], "category": "b2b_sales", "quality_tier": "verified", "requires_env": [], "dependencies": [ "@rapp/basic-agent" ], "_file": "agents/@aibast-agents-library/b2b_sales_stacks/deal_progression_stack/activity_gap_agent.py", "_size_kb": 18.0, "_lines": 393, "_sha256": "e78373cf5e31d61e079c15f6562294c9cbf96cae4dc738ca2cfabbd50f050f59", "_install_prefix": "activity_gap__", "_install_filename": "activity_gap__e78373cf5e31_agent.py", "_stack": "deal_progression", "_stack_vertical": "b2b_sales", "_synthetic_data": true, "_added_at": "2026-03-18T13:23:12-04:00", "_catalog_kind": "component", "_readiness": [ "Synthetic data" ] }, { "schema": "rapp-agent/1.0", "name": "@aibast-agents-library/deal-competitor-intel", "version": "1.0.0", "display_name": "Deal Competitor Intelligence", "description": "Competitive snapshots, threat scoring, counter-strategies, and win/loss pattern analysis per deal.", "author": "AIBAST", "tags": [ "b2b", "sales", "competitive-intelligence", "deal-progression", "strategy" ], "category": "b2b_sales", "quality_tier": "verified", "requires_env": [], "dependencies": [ "@rapp/basic-agent" ], "_file": "agents/@aibast-agents-library/b2b_sales_stacks/deal_progression_stack/competitor_intelligence_agent.py", "_size_kb": 16.5, "_lines": 334, "_sha256": "c80a6d56f47d8f7c1cdba0dbe6b7fff580e5dccef2cfb71ed0e38086401a32d7", "_install_prefix": "deal_competitor_intel__", "_install_filename": "deal_competitor_intel__c80a6d56f47d_agent.py", "_stack": "deal_progression", "_stack_vertical": "b2b_sales", "_synthetic_data": true, "_added_at": "2026-03-18T13:23:12-04:00", "_catalog_kind": "component", "_readiness": [ "Synthetic data" ] }, { "schema": "rapp-agent/1.0", "name": "@aibast-agents-library/deal-health-score", "version": "1.0.0", "display_name": "Deal Health Score", "description": "Calculates deal health scores, trend analysis, benchmark comparison, and health alerts.", "author": "AIBAST", "tags": [ "b2b", "sales", "deal-health", "scoring", "pipeline" ], "category": "b2b_sales", "quality_tier": "verified", "requires_env": [], "dependencies": [ "@rapp/basic-agent" ], "_file": "agents/@aibast-agents-library/b2b_sales_stacks/deal_progression_stack/deal_health_score_agent.py", "_size_kb": 17.6, "_lines": 378, "_sha256": "dc4353257490668d9b07c0a8433831ed1bf28f9f01be62b3789f12d1d9b87839", "_install_prefix": "deal_health_score__", "_install_filename": "deal_health_score__dc4353257490_agent.py", "_stack": "deal_progression", "_stack_vertical": "b2b_sales", "_synthetic_data": true, "_added_at": "2026-03-18T13:23:12-04:00", "_catalog_kind": "component", "_readiness": [ "Synthetic data" ] }, { "schema": "rapp-agent/1.0", "name": "@aibast-agents-library/deal-progression", "version": "1.0.0", "display_name": "Deal Progression Agent", "description": "Automate sales pipeline management to keep deals moving, increase forecast confidence, and improve team productivity.", "author": "AIBAST", "tags": [ "b2b", "sales", "deal-progression", "pipeline", "forecasting" ], "category": "b2b_sales", "quality_tier": "verified", "requires_env": [], "dependencies": [ "@rapp/basic-agent" ], "_file": "agents/@aibast-agents-library/b2b_sales_stacks/deal_progression_stack/deal_progression_agent.py", "_size_kb": 46.3, "_lines": 684, "_sha256": "7ab701819a48cd143e6047afd00f303bbbbc32a42ae1dd6b81c60181379818ce", "_install_prefix": "deal_progression__", "_install_filename": "deal_progression__7ab701819a48_agent.py", "_stack": "deal_progression", "_stack_vertical": "b2b_sales", "_synthetic_data": true, "_added_at": "2026-03-18T13:23:12-04:00", "_solution": { "advertised_name": "Deal Progression Agent", "sharepoint_list_name": "Sales Pipeline Management Agent", "slot": 12, "executive_summary": "Automate sales pipeline management to keep deals moving, increase forecast confidence, and improve team productivity.", "industries": [ "Cross-Industry" ], "personas": [ "Account Executive", "Sales Director" ], "featured_tools": [ "Dynamics 365", "SharePoint", "Microsoft Teams" ], "agent_requirements": [ "Microsoft Copilot Studio" ], "capabilities": [ "Get a real-time pipeline assessment", "Identify stalled deals, then run gap analysis and stakeholder tracking", "Diagnose the exact blockers to deal momentum like legal delays", "Generate tailored action plans with tasks, deadlines, and talk tracks for each stalled opportunity" ], "outcomes": [ "Protected their pipeline health, reducing at-risk deals", "Accelerated deal momentum with shorter sales cycles", "Improved forecast confidence and reduced slippage", "Reduced hours of manual review, enabling managers to focus on high-impact opportunities" ], "customer_scenario": [ "Forecast accuracy was at risk due to lack of visibility into the root causes of deal slippage", "Manual pipeline reviews took hours without offering actionable insights", "Managers lacked visibility into engagement gaps and competitive threats" ], "slide_title": "Deal Progression Agent", "slide_summary": "Automate sales pipeline management to keep deals moving, increase forecast confidence, and improve team productivity.", "scenario_name": "A SaaS company", "challenge_intro": "A regional sales team faced quarter-end pressure with multiple high-value deals stalled in late stages:", "actions_intro": "The team used the Deal Progression Agent to:", "outcomes_intro": "With the Agent, the team:", "opportunity_statements": [ "Perform proactive pipeline analysis to improve forecast confidence", "Detect stalled deals to reduce slippage and protect revenue commitments", "Diagnose why deals lost momentum to enable more targeted interventions", "Generate action plans that accelerate deal progress and improve close rates" ], "module_name": null, "module_description": null, "source_verified": true, "sales_headline": "Turn pipeline review into a focused intervention plan", "card_pitch": "Give account executives and sales directors a shared view of pipeline health, stalled-deal evidence, blockers, and reviewable next steps without automating CRM actions.", "why_try": "A leader can move from a broad pipeline question to the specific opportunities, engagement gaps, and candidate interventions that deserve human attention.", "customer_challenge_copy": "Sales teams spend too much time reconciling pipeline views and still lack consistent evidence for why late-stage deals have slowed.", "microsoft_ai_story": "Microsoft Copilot Studio provides the conversational review surface. Dynamics 365 Sales supplies opportunity and activity context, SharePoint holds approved playbooks, and Microsoft Teams supports manager-led follow-up after review.", "business_value": [ "Improves visibility into pipeline health and stalled-deal evidence.", "Makes blocker diagnosis and intervention planning more consistent.", "Supports forecast review while keeping commitments and CRM actions human governed." ], "search_terms": [ "deal progression", "pipeline health", "stalled deals", "deal blockers", "sales forecast review", "pipeline velocity", "next best action", "Dynamics 365 Sales", "SharePoint sales playbooks", "Microsoft Teams sales", "account executive", "Copilot Studio" ], "journey_stage": "Optimize", "blueprint_role": "Provides a read-only pipeline intelligence layer that turns synthetic activity and stage evidence into reviewable intervention options.", "sample_prompts": [ { "label": "Review pipeline health", "prompt": "Which opportunities need attention in the synthetic pipeline, and what evidence should I review before changing the forecast?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=deal-progression-pipeline-health" }, { "label": "Diagnose stalled deals", "prompt": "What blocker evidence explains the loss of momentum on our two largest stalled synthetic deals?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=deal-progression-stalled-deals" }, { "label": "Draft intervention plans", "prompt": "Draft reviewable intervention plans for our two largest stalled synthetic deals, but do not assign work or contact anyone.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=deal-progression-action-plans" }, { "label": "Compare acceleration options", "prompt": "Which synthetic timing options could move pipeline review forward without turning scenario value into a forecast commitment?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=deal-progression-acceleration" }, { "label": "Map candidate tasks", "prompt": "Map candidate follow-up work to the synthetic rep capacity for my review; do not create tasks or alerts.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=deal-progression-assign-tasks" }, { "label": "Prepare the leadership summary", "prompt": "Give me a leadership-ready summary of the synthetic pipeline findings and the decisions that still require human review.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=deal-progression-executive-summary" } ], "architecture": { "business_flow": [ "Account executive or sales director", "Copilot experience", "Deal Progression Agent", "Synthetic pipeline evidence", "Human pipeline review" ], "capabilities": [ { "operation": "pipeline_health", "name": "Pipeline Health", "purpose": "Classifies active opportunities and surfaces the pipeline areas that need review." }, { "operation": "stalled_deals", "name": "Stalled Deals", "purpose": "Explains stalled opportunities using stage age, engagement, champion, and blocker evidence." }, { "operation": "action_plans", "name": "Action Plans", "purpose": "Drafts candidate interventions, owners, and timing for human review." }, { "operation": "acceleration", "name": "Acceleration", "purpose": "Compares timing options without turning scenario value into a forecast commitment." }, { "operation": "assign_tasks", "name": "Assign Tasks", "purpose": "Maps candidate tasks to rep capacity without creating assignments or alerts." }, { "operation": "executive_summary", "name": "Executive Summary", "purpose": "Condenses findings and draft next steps for leadership review." } ], "easy_mode": [ "Verify Power Platform CLI and authenticate to the target environment.", "Initialize a modern cli-copilot project with an approved publisher prefix.", "Upload the two synthetic knowledge files and create one manual skill for every advertised operation.", "Bind dedicated approved connection references only after replacing the synthetic evidence route.", "Push the draft, replay every locked persona case, and stop before publication or any external action." ], "local_install_prompt": "You are GitHub Copilot running in Agent mode in VS Code. Install and open the AI BAST Deal Progression Agent from this deployment recipe: https://raw.githubusercontent.com/microsoft/aibast-agents-library/main/solutions/deal-progression/deployment.json. Own the setup and do not ask me to run terminal commands. Inspect the operating system; verify the Brainstem health endpoint; if needed, use the official installer from the recipe; fetch and install the portable source; verify the expected tool; and replay the recipe smoke test. Preserve the synthetic-only route for pipeline_health, stalled_deals, action_plans, acceleration, assign_tasks, executive_summary. Do not connect customer systems, perform writes, publish, approve pricing, send outreach, or silently substitute another agent. If authentication requires browser confirmation, ask only for that browser step. Stop and report the exact evidence and any failed step. Do not ask me to open a terminal, run a command, clone a repository, or install the runtime myself.", "copilot_studio_prompt": "You are GitHub Copilot running in Agent mode in VS Code with the Microsoft Copilot Studio plugin installed. Prepare the AI BAST Deal Progression Agent from https://raw.githubusercontent.com/microsoft/aibast-agents-library/main/solutions/deal-progression/deployment.json. Ask only for the target Power Platform environment ID and publisher prefix. Verify PAC CLI, authenticate if needed, initialize a cli-copilot project, upload the two synthetic knowledge files, and create one skill for each operation: pipeline_health, stalled_deals, action_plans, acceleration, assign_tasks, executive_summary. Bind only dedicated approved connection references, preserve every evidence and approval boundary, pull before push, and replay every locked persona case. Stop before publish and show the generated files, validation evidence, unresolved connection work, and the exact publish command that would require user approval.", "required_connections": [ "Dynamics 365 Sales", "SharePoint", "Microsoft Teams" ], "manual_commands": [ "pac auth create", "pac copilot init --name \"Deal Progression Agent\" --publisher-prefix --authoring-mode cli-copilot --project-dir \"\" --environment \"\"", "pac connection list --environment \"\"", "pac copilot pull --project-dir \"\"", "pac copilot push --project-dir \"\"", "pac copilot publish --bot \"\" --environment \"\"" ], "acceptance_checks": [ "The package remains under solutions/deal-progression and the portable source resolves to the expected tool.", "Every advertised operation has one uploadable skill and one locked persona-language capture case.", "The approved one-pager filename and source hash remain pinned in the one-pager map.", "Every exact value is identified as synthetic or as a planning assumption rather than a measured customer outcome.", "No CRM write, outreach, approval, pricing change, forecast change, task, alert, document delivery, or customer communication occurs.", "Publishing remains a separate user-approved step after push and validation." ], "hard_mode": [ "Run pac copilot init and confirm settings.mcs.yml preserves the intended environment, language, and publisher identity.", "Author global instructions that route only the supported operations: pipeline_health, stalled_deals, action_plans, acceleration, assign_tasks, executive_summary.", "Upload both synthetic knowledge files and create each operation skill from the package SKILL.md source.", "Create dedicated connection references for the approved production seams; do not reuse ambiguous existing connections.", "Validate schema, instructions, skills, knowledge, and connection bindings; pull before push and replay every locked case.", "Keep the agent in Draft and require explicit confirmation before publish or any external action." ] }, "curated_copy": true, "source": "catalog", "origin": null, "onepager": "Deal Progression Agent one-pager.pptx", "demo_video": "15-Deal Progression Agent.mp4", "has_onepager": true, "has_demo_video": true, "promise_coverage": 70, "is_primary": true, "aliases": [], "package": { "slug": "deal-progression", "quest_url": "solutions/deal-progression/quest.html", "manual_tutorial_url": "solutions/deal-progression/manual-tutorial.html", "export_manifest_url": "solutions/deal-progression/export-manifest.json" } }, "_demo": { "slug": "deal-progression", "case_count": 6, "personas": [ "Account Executive", "Sales Director" ], "prompts": [ "Which opportunities need attention in the synthetic pipeline, and what evidence should I review before changing the forecast?", "What blocker evidence explains the loss of momentum on our two largest stalled synthetic deals?", "Draft reviewable intervention plans for our two largest stalled synthetic deals, but do not assign work or contact anyone." ] }, "_catalog_kind": "solution", "_readiness": [ "Demo proven", "Conversation tested", "Synthetic data", "Production pattern" ] }, { "schema": "rapp-agent/1.0", "name": "@aibast-agents-library/deal-risk-assessment", "version": "1.0.0", "display_name": "Deal Risk Assessment", "description": "Multi-factor risk scoring, risk matrices, mitigation planning, and risk trend tracking for pipeline deals.", "author": "AIBAST", "tags": [ "b2b", "sales", "risk-assessment", "deal-progression", "pipeline" ], "category": "b2b_sales", "quality_tier": "verified", "requires_env": [], "dependencies": [ "@rapp/basic-agent" ], "_file": "agents/@aibast-agents-library/b2b_sales_stacks/deal_progression_stack/deal_risk_assessment_agent.py", "_size_kb": 19.2, "_lines": 424, "_sha256": "7f5aa8a7cdf7db08e34903b432ab1aae93ff4c69a53f5cd74f14d0069f9ff77d", "_install_prefix": "deal_risk_assessment__", "_install_filename": "deal_risk_assessment__7f5aa8a7cdf7_agent.py", "_stack": "deal_progression", "_stack_vertical": "b2b_sales", "_synthetic_data": true, "_added_at": "2026-03-18T13:23:12-04:00", "_catalog_kind": "component", "_readiness": [ "Synthetic data" ] }, { "schema": "rapp-agent/1.0", "name": "@aibast-agents-library/next-best-action", "version": "1.0.0", "display_name": "Next Best Action", "description": "Recommends prioritized actions, sequences tasks, forecasts impact, and optimizes rep assignments.", "author": "AIBAST", "tags": [ "b2b", "sales", "next-best-action", "deal-progression", "recommendations" ], "category": "b2b_sales", "quality_tier": "verified", "requires_env": [], "dependencies": [ "@rapp/basic-agent" ], "_file": "agents/@aibast-agents-library/b2b_sales_stacks/deal_progression_stack/next_best_action_agent.py", "_size_kb": 19.7, "_lines": 430, "_sha256": "34e511cdf408effb22f5e3b377e2a82201881c416371908b9db58fef307a3fdb", "_install_prefix": "next_best_action__", "_install_filename": "next_best_action__34e511cdf408_agent.py", "_stack": "deal_progression", "_stack_vertical": "b2b_sales", "_synthetic_data": true, "_added_at": "2026-03-18T13:23:12-04:00", "_catalog_kind": "component", "_readiness": [ "Synthetic data" ] }, { "schema": "rapp-agent/1.0", "name": "@aibast-agents-library/pipeline-velocity", "version": "1.0.0", "display_name": "Pipeline Velocity", "description": "Velocity dashboard, stage analysis, bottleneck detection, and acceleration planning for the sales pipeline.", "author": "AIBAST", "tags": [ "b2b", "sales", "pipeline-velocity", "deal-progression", "analytics" ], "category": "b2b_sales", "quality_tier": "verified", "requires_env": [], "dependencies": [ "@rapp/basic-agent" ], "_file": "agents/@aibast-agents-library/b2b_sales_stacks/deal_progression_stack/pipeline_velocity_agent.py", "_size_kb": 17.1, "_lines": 380, "_sha256": "639bf408848cdaf12398fdaf98c8c2029f80ce87ecbae34169e6a6770c33fbe6", "_install_prefix": "pipeline_velocity__", "_install_filename": "pipeline_velocity__639bf408848c_agent.py", "_stack": "deal_progression", "_stack_vertical": "b2b_sales", "_synthetic_data": true, "_added_at": "2026-03-18T13:23:12-04:00", "_catalog_kind": "component", "_readiness": [ "Synthetic data" ] }, { "schema": "rapp-agent/1.0", "name": "@aibast-agents-library/revenue-forecast", "version": "1.0.0", "display_name": "Revenue Forecast", "description": "Quarterly forecasting, scenario analysis, commit vs best-case modeling, and forecast accuracy tracking.", "author": "AIBAST", "tags": [ "b2b", "sales", "revenue-forecast", "deal-progression", "analytics" ], "category": "b2b_sales", "quality_tier": "verified", "requires_env": [], "dependencies": [ "@rapp/basic-agent" ], "_file": "agents/@aibast-agents-library/b2b_sales_stacks/deal_progression_stack/revenue_forecast_agent.py", "_size_kb": 16.1, "_lines": 348, "_sha256": "3a9f7b8165a071e91403c7bff947b7caf395d1772b51a185fcbb26ab3d7f41ba", "_install_prefix": "revenue_forecast__", "_install_filename": "revenue_forecast__3a9f7b8165a0_agent.py", "_stack": "deal_progression", "_stack_vertical": "b2b_sales", "_synthetic_data": true, "_added_at": "2026-03-18T13:23:12-04:00", "_catalog_kind": "component", "_readiness": [ "Synthetic data" ] }, { "schema": "rapp-agent/1.0", "name": "@aibast-agents-library/deal-stakeholder-engagement", "version": "1.0.0", "display_name": "Stakeholder Engagement", "description": "Engagement scoring, relationship mapping, engagement planning, and sentiment analysis for deal stakeholders.", "author": "AIBAST", "tags": [ "b2b", "sales", "stakeholder-engagement", "deal-progression", "relationships" ], "category": "b2b_sales", "quality_tier": "verified", "requires_env": [], "dependencies": [ "@rapp/basic-agent" ], "_file": "agents/@aibast-agents-library/b2b_sales_stacks/deal_progression_stack/stakeholder_engagement_agent.py", "_size_kb": 19.2, "_lines": 398, "_sha256": "7fb6cd8613c6f9de8748448fc06aed54fbd5aed3bf1ac828022a09f56bca7394", "_install_prefix": "deal_stakeholder_engagement__", "_install_filename": "deal_stakeholder_engagement__7fb6cd8613c6_agent.py", "_stack": "deal_progression", "_stack_vertical": "b2b_sales", "_synthetic_data": true, "_added_at": "2026-03-18T13:23:12-04:00", "_catalog_kind": "component", "_readiness": [ "Synthetic data" ] }, { "schema": "rapp-agent/1.0", "name": "@aibast-agents-library/stalled-deal-detection", "version": "1.0.0", "display_name": "Stalled Deal Detection", "description": "Detects stalled deals, classifies root causes, generates intervention plans, and provides stall prevention.", "author": "AIBAST", "tags": [ "b2b", "sales", "stalled-deals", "deal-progression", "pipeline" ], "category": "b2b_sales", "quality_tier": "verified", "requires_env": [], "dependencies": [ "@rapp/basic-agent" ], "_file": "agents/@aibast-agents-library/b2b_sales_stacks/deal_progression_stack/stalled_deal_detection_agent.py", "_size_kb": 20.6, "_lines": 453, "_sha256": "e05038240707d856c23934051fd47b2c3ad6260120902de43f010ec0ed776929", "_install_prefix": "stalled_deal_detection__", "_install_filename": "stalled_deal_detection__e05038240707_agent.py", "_stack": "deal_progression", "_stack_vertical": "b2b_sales", "_synthetic_data": true, "_added_at": "2026-03-18T13:23:12-04:00", "_catalog_kind": "component", "_readiness": [ "Synthetic data" ] }, { "schema": "rapp-agent/1.0", "name": "@aibast-agents-library/win-probability", "version": "1.0.0", "display_name": "Win Probability", "description": "Calculates win probabilities, analyzes contributing factors, tracks trends, and compares deals.", "author": "AIBAST", "tags": [ "b2b", "sales", "win-probability", "deal-progression", "forecasting" ], "category": "b2b_sales", "quality_tier": "verified", "requires_env": [], "dependencies": [ "@rapp/basic-agent" ], "_file": "agents/@aibast-agents-library/b2b_sales_stacks/deal_progression_stack/win_probability_agent.py", "_size_kb": 20.4, "_lines": 414, "_sha256": "c75e86c4dc900bde852af7ceab38c07be15a18f5ac96d999c6f3999e5bb6384a", "_install_prefix": "win_probability__", "_install_filename": "win_probability__c75e86c4dc90_agent.py", "_stack": "deal_progression", "_stack_vertical": "b2b_sales", "_synthetic_data": true, "_added_at": "2026-03-18T13:23:12-04:00", "_catalog_kind": "component", "_readiness": [ "Synthetic data" ] }, { "schema": "rapp-agent/1.0", "name": "@aibast-agents-library/proposal-generation", "version": "1.0.0", "display_name": "Proposal Generation Agent", "description": "Automate proposal creation to accelerate deal cycles, improve win rates, and deliver consistent, high-quality responses.", "author": "AIBAST", "tags": [ "b2b", "sales", "proposal", "rfp", "pricing", "competitive-positioning" ], "category": "b2b_sales", "quality_tier": "verified", "requires_env": [], "dependencies": [ "@rapp/basic-agent" ], "_file": "agents/@aibast-agents-library/b2b_sales_stacks/proposal_generation_stack/proposal_generation_agent.py", "_size_kb": 37.4, "_lines": 663, "_sha256": "c6fd0c72321522331ef8c5de317bd236f3be7f1ac59936cade4f31289f52a6b5", "_install_prefix": "proposal_generation__", "_install_filename": "proposal_generation__c6fd0c723215_agent.py", "_stack": "proposal_generation", "_stack_vertical": "b2b_sales", "_synthetic_data": true, "_added_at": "2026-03-18T13:23:12-04:00", "_solution": { "advertised_name": "Proposal Generation Agent", "sharepoint_list_name": "Proposal Creation Agent", "slot": 13, "executive_summary": "Automate proposal creation to accelerate deal cycles, improve win rates, and deliver consistent, high-quality responses.", "industries": [ "Cross-Industry" ], "personas": [ "Account Executive", "Sales Leader", "Bid Manager" ], "featured_tools": [ "Dynamics 365", "Microsoft Teams", "Word / PowerPoint" ], "agent_requirements": [ "Microsoft Copilot Studio" ], "capabilities": [ "Parse RFP requirements and draft compliance checklists instantly", "Assemble tailored executive summaries and competitive positioning", "Optimize pricing and bundle recommendations", "Draft client-ready proposal materials" ], "outcomes": [ "Developed a complete, branded proposal package in minutes", "Improved proposal quality and alignment with buyer priorities", "Enabled same-day turnaround for competitive opportunities", "Freed sales teams to pursue more deals without sacrificing proposal quality" ], "customer_scenario": [ "Manual proposal creation consumed hours, risking missed deadlines", "Sellers lacked visibility into compliance gaps and stakeholder priorities", "Inconsistent planning and content quality hindered competitive positioning" ], "slide_title": "Proposal Generation Agent", "slide_summary": "Automate proposal creation to accelerate deal cycles, improve win rates, and deliver consistent, high-quality responses.", "scenario_name": "An information technology company", "challenge_intro": "A technology company faced an urgent RFP deadline for a high-value healthcare project:", "actions_intro": "The company leveraged the Proposal Generation Agent to:", "outcomes_intro": "With the agent\u2019s help, the company:", "opportunity_statements": [ "Perform rapid RFP analysis and requirement extraction to reduce risk and improve accuracy", "Assemble personalized proposal content and positioning to strengthen differentiation and win probability", "Generate branded, professional proposal materials for same-day delivery to shorten sales cycles" ], "module_name": null, "module_description": null, "source_verified": true, "sales_headline": "Move from RFP requirements to a reviewable proposal structure", "card_pitch": "Help bid teams extract requirements, tailor positioning, compare pricing assumptions, and prepare a proposal package outline while preserving legal, pricing, reference, and delivery approvals.", "why_try": "A bid manager can see how buyer requirements map to evidence and where human review is still required before anything becomes customer ready.", "customer_challenge_copy": "Urgent RFP work is fragmented across requirement analysis, reusable content, pricing, references, and approvals, creating inconsistency and avoidable rework.", "microsoft_ai_story": "Microsoft Copilot Studio guides the proposal workflow. Dynamics 365 Sales contributes opportunity context, SharePoint provides approved response content, Microsoft Word and PowerPoint provide document surfaces, and Microsoft Teams supports bid-team review.", "business_value": [ "Improves traceability from stated requirements to draft response content.", "Creates a consistent structure for positioning, pricing review, and supporting evidence.", "Keeps proposal delivery and commercial approvals under authorized human control." ], "search_terms": [ "proposal generation", "RFP analysis", "requirement extraction", "executive summary", "proposal pricing", "competitive positioning", "bid management", "Dynamics 365 Sales", "SharePoint content library", "Microsoft Word proposals", "Microsoft PowerPoint", "Copilot Studio" ], "journey_stage": "Create", "blueprint_role": "Establishes a governed proposal-preparation workflow from requirement extraction through a reviewable delivery outline.", "sample_prompts": [ { "label": "Analyze the RFP", "prompt": "Analyze the synthetic Meridian Healthcare RFP and show the traceable requirement checklist.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=proposal-generation-analyze-rfp" }, { "label": "Draft the executive summary", "prompt": "Draft an executive summary for the synthetic Meridian Healthcare opportunity that reflects the buyer priorities and remains subject to review.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=proposal-generation-executive-summary" }, { "label": "Review pricing assumptions", "prompt": "Compare the synthetic solution and pricing assumptions for Meridian Healthcare without approving a price, discount, or concession.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=proposal-generation-solution-pricing" }, { "label": "Prepare competitive positioning", "prompt": "Prepare synthetic reference and competitive positioning options for Meridian Healthcare, with availability checks before use.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=proposal-generation-references-positioning" }, { "label": "Outline the proposal package", "prompt": "Outline the synthetic Meridian Healthcare proposal package and every human review required before delivery.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=proposal-generation-compile-proposal" }, { "label": "Summarize draft readiness", "prompt": "Summarize the synthetic Meridian Healthcare draft readiness and the decisions authorized reviewers must make next.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=proposal-generation-delivery-summary" } ], "architecture": { "business_flow": [ "Account executive, sales leader, or bid manager", "Copilot experience", "Proposal Generation Agent", "Synthetic RFP and content evidence", "Authorized proposal review" ], "capabilities": [ { "operation": "analyze_rfp", "name": "Analyze Rfp", "purpose": "Extracts and maps requirements to synthetic capability evidence." }, { "operation": "executive_summary", "name": "Executive Summary", "purpose": "Drafts buyer-aligned executive positioning for review." }, { "operation": "solution_pricing", "name": "Solution Pricing", "purpose": "Compares solution, budget, discount, and margin assumptions without approval." }, { "operation": "references_positioning", "name": "References Positioning", "purpose": "Drafts reference and competitor positioning while requiring availability validation." }, { "operation": "compile_proposal", "name": "Compile Proposal", "purpose": "Builds a package outline and required human-review checklist." }, { "operation": "delivery_summary", "name": "Delivery Summary", "purpose": "Summarizes draft readiness and human-governed next-step options." } ], "easy_mode": [ "Verify Power Platform CLI and authenticate to the target environment.", "Initialize a modern cli-copilot project with an approved publisher prefix.", "Upload the two synthetic knowledge files and create one manual skill for every advertised operation.", "Bind dedicated approved connection references only after replacing the synthetic evidence route.", "Push the draft, replay every locked persona case, and stop before publication or any external action." ], "local_install_prompt": "You are GitHub Copilot running in Agent mode in VS Code. Install and open the AI BAST Proposal Generation Agent from this deployment recipe: https://raw.githubusercontent.com/microsoft/aibast-agents-library/main/solutions/proposal-generation/deployment.json. Own the setup and do not ask me to run terminal commands. Inspect the operating system; verify the Brainstem health endpoint; if needed, use the official installer from the recipe; fetch and install the portable source; verify the expected tool; and replay the recipe smoke test. Preserve the synthetic-only route for analyze_rfp, executive_summary, solution_pricing, references_positioning, compile_proposal, delivery_summary. Do not connect customer systems, perform writes, publish, approve pricing, send outreach, or silently substitute another agent. If authentication requires browser confirmation, ask only for that browser step. Stop and report the exact evidence and any failed step. Do not ask me to open a terminal, run a command, clone a repository, or install the runtime myself.", "copilot_studio_prompt": "You are GitHub Copilot running in Agent mode in VS Code with the Microsoft Copilot Studio plugin installed. Prepare the AI BAST Proposal Generation Agent from https://raw.githubusercontent.com/microsoft/aibast-agents-library/main/solutions/proposal-generation/deployment.json. Ask only for the target Power Platform environment ID and publisher prefix. Verify PAC CLI, authenticate if needed, initialize a cli-copilot project, upload the two synthetic knowledge files, and create one skill for each operation: analyze_rfp, executive_summary, solution_pricing, references_positioning, compile_proposal, delivery_summary. Bind only dedicated approved connection references, preserve every evidence and approval boundary, pull before push, and replay every locked persona case. Stop before publish and show the generated files, validation evidence, unresolved connection work, and the exact publish command that would require user approval.", "required_connections": [ "Dynamics 365 Sales", "SharePoint", "Microsoft Teams", "Microsoft Word or PowerPoint document workflow", "Authorized pricing approval workflow" ], "manual_commands": [ "pac auth create", "pac copilot init --name \"Proposal Generation Agent\" --publisher-prefix --authoring-mode cli-copilot --project-dir \"\" --environment \"\"", "pac connection list --environment \"\"", "pac copilot pull --project-dir \"\"", "pac copilot push --project-dir \"\"", "pac copilot publish --bot \"\" --environment \"\"" ], "acceptance_checks": [ "The package remains under solutions/proposal-generation and the portable source resolves to the expected tool.", "Every advertised operation has one uploadable skill and one locked persona-language capture case.", "The approved one-pager filename and source hash remain pinned in the one-pager map.", "Every exact value is identified as synthetic or as a planning assumption rather than a measured customer outcome.", "No CRM write, outreach, approval, pricing change, forecast change, task, alert, document delivery, or customer communication occurs.", "Publishing remains a separate user-approved step after push and validation." ], "hard_mode": [ "Run pac copilot init and confirm settings.mcs.yml preserves the intended environment, language, and publisher identity.", "Author global instructions that route only the supported operations: analyze_rfp, executive_summary, solution_pricing, references_positioning, compile_proposal, delivery_summary.", "Upload both synthetic knowledge files and create each operation skill from the package SKILL.md source.", "Create dedicated connection references for the approved production seams; do not reuse ambiguous existing connections.", "Validate schema, instructions, skills, knowledge, and connection bindings; pull before push and replay every locked case.", "Keep the agent in Draft and require explicit confirmation before publish or any external action." ] }, "curated_copy": true, "source": "catalog", "origin": null, "onepager": "Proposal Generation Agent one-pager.pptx", "demo_video": "20-Proposal Generation Agent.mp4", "has_onepager": true, "has_demo_video": true, "promise_coverage": 65, "is_primary": true, "aliases": [], "package": { "slug": "proposal-generation", "quest_url": "solutions/proposal-generation/quest.html", "manual_tutorial_url": "solutions/proposal-generation/manual-tutorial.html", "export_manifest_url": "solutions/proposal-generation/export-manifest.json" } }, "_demo": { "slug": "proposal-generation", "case_count": 6, "personas": [ "Account Executive", "Bid Manager", "Sales Leader" ], "prompts": [ "Analyze the synthetic Meridian Healthcare RFP and show the traceable requirement checklist.", "Draft an executive summary for the synthetic Meridian Healthcare opportunity that reflects the buyer priorities and remains subject to review.", "Compare the synthetic solution and pricing assumptions for Meridian Healthcare without approving a price, discount, or concession." ] }, "_catalog_kind": "solution", "_readiness": [ "Demo proven", "Conversation tested", "Synthetic data", "Production pattern" ] }, { "schema": "rapp-agent/1.0", "name": "@aibast-agents-library/sales-qualification", "version": "1.0.0", "display_name": "Sales Qualification Agent", "description": "Automate lead scoring and qualification to accelerate pipeline generation, improve conversion rates, and drive proactive sales.", "author": "AIBAST", "tags": [ "b2b", "sales", "lead-qualification", "bant", "icp-scoring", "lead-routing" ], "category": "b2b_sales", "quality_tier": "verified", "requires_env": [], "dependencies": [ "@rapp/basic-agent" ], "_file": "agents/@aibast-agents-library/b2b_sales_stacks/sales_qualification_stack/sales_qualification_agent.py", "_size_kb": 51.6, "_lines": 551, "_sha256": "5ec782d1e5ef9906de2e37a624c238933408b3fa866ffa25e5194eb46ad75141", "_install_prefix": "sales_qualification__", "_install_filename": "sales_qualification__5ec782d1e5ef_agent.py", "_stack": "sales_qualification", "_stack_vertical": "b2b_sales", "_synthetic_data": true, "_added_at": "2026-03-18T13:23:12-04:00", "_solution": { "advertised_name": "Sales Qualification Agent", "sharepoint_list_name": "Sales Qualification Agent", "slot": 21, "executive_summary": "Automate lead scoring and qualification to accelerate pipeline generation, improve conversion rates, and drive proactive sales.", "industries": [], "personas": [], "featured_tools": [ "Dynamics 365", "Microsoft Teams" ], "agent_requirements": [ "Microsoft Copilot Studio" ], "capabilities": [ "Enrich leads with firmographic, technographic, and intent data", "Score and segment leads using ICP and BANT criteria for prioritization", "Generate tailored outreach sequences and talking points", "Automate SLA tracking and alerts to ensure timely follow-up" ], "outcomes": [ "Reduced lead qualification time from hours to minutes", "Increased lead-to-meeting conversion through personalized engagement", "Balanced workload across sales reps while maintaining SLAs", "Freed team capacity to focus on strategic, high-value opportunities" ], "customer_scenario": [ "Manual qualification consumed hours and risked lost opportunities", "Sales reps lacked complete context for personalized outreach", "Lead routing was inconsistent, creating workload imbalances", "Leadership had limited visibility into SLA compliance for hot leads" ], "slide_title": "Sales Qualification Agent", "slide_summary": "Automate lead scoring and qualification to accelerate pipeline generation, improve conversion rates, and drive proactive sales.", "scenario_name": "A SaaS company", "challenge_intro": "A SaaS company faced a surge of inbound leads after a trade show:", "actions_intro": "The SaaS company leveraged the Sales Qualification Agent to:", "outcomes_intro": "With the agent\u2019s help, the company:", "opportunity_statements": [ "Perform intelligent lead enrichment to prioritize high-potential opportunities", "Automate qualification scoring to identify conversion-ready prospects", "Generate personalized outreach sequences to boost engagement", "Coordinate SLA-based assignments and tracking to drive timely engagement" ], "module_name": null, "module_description": null, "source_verified": true, "sales_headline": "Make lead qualification consistent before outreach begins", "card_pitch": "Help sales managers and business-development teams review enrichment, ICP fit, BANT evidence, draft outreach, routing recommendations, and SLA plans from one governed workflow.", "why_try": "A team can see why a lead is prioritized, what qualification evidence is missing, and what should be reviewed next without triggering outreach or CRM automation.", "customer_challenge_copy": "High lead volume makes manual qualification inconsistent, while incomplete context and uneven routing create avoidable follow-up risk.", "microsoft_ai_story": "Microsoft Copilot Studio gives sellers a qualification workspace. Dynamics 365 Sales supplies lead and ownership context, approved enrichment connectors contribute firmographic and intent evidence, and Microsoft Teams supports manager review of routing and SLA plans.", "business_value": [ "Makes qualification criteria and evidence easier to inspect.", "Creates consistent draft outreach and routing recommendations.", "Supports SLA planning while preserving seller and manager control." ], "search_terms": [ "sales qualification", "lead scoring", "ICP scoring", "BANT qualification", "lead enrichment", "lead routing", "SLA tracking", "Dynamics 365 Sales leads", "intent data connector", "Microsoft Teams sales", "business development", "Copilot Studio" ], "journey_stage": "Prioritize", "blueprint_role": "Provides a governed qualification layer between inbound lead evidence and authorized sales engagement.", "sample_prompts": [ { "label": "Prioritize synthetic leads", "prompt": "Which bundled synthetic leads should my team review first, and why did they score that way?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=sales-qualification-score-leads" }, { "label": "Review BANT evidence", "prompt": "Show the BANT evidence and missing qualification details for the strongest synthetic leads.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=sales-qualification-bant-analysis" }, { "label": "Draft outreach ideas", "prompt": "Draft outreach ideas for the synthetic hot leads, but do not send or schedule any communication.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=sales-qualification-create-outreach" }, { "label": "Recommend lead routing", "prompt": "Recommend synthetic lead routing for manager review without assigning CRM owners.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=sales-qualification-assign-leads" }, { "label": "Draft the SLA plan", "prompt": "Draft an SLA and escalation plan for the synthetic leads without activating alerts or automations.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=sales-qualification-setup-tracking" }, { "label": "Summarize qualification evidence", "prompt": "Summarize the synthetic qualified pipeline and clearly label every conversion and value assumption.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=sales-qualification-qualification-report" } ], "architecture": { "business_flow": [ "Business development rep, account executive, or sales manager", "Copilot experience", "Sales Qualification Agent", "Synthetic lead and capacity evidence", "Human qualification and routing review" ], "capabilities": [ { "operation": "score_leads", "name": "Score Leads", "purpose": "Applies the documented synthetic ICP and tiering model." }, { "operation": "bant_analysis", "name": "Bant Analysis", "purpose": "Breaks down budget, authority, need, and timeline evidence." }, { "operation": "create_outreach", "name": "Create Outreach", "purpose": "Drafts reviewable outreach ideas without sending or scheduling them." }, { "operation": "assign_leads", "name": "Assign Leads", "purpose": "Recommends routing from territory, specialty, and capacity assumptions." }, { "operation": "setup_tracking", "name": "Setup Tracking", "purpose": "Drafts SLA and escalation rules without activating alerts." }, { "operation": "qualification_report", "name": "Qualification Report", "purpose": "Summarizes qualification evidence and review actions without conversion commitments." } ], "easy_mode": [ "Verify Power Platform CLI and authenticate to the target environment.", "Initialize a modern cli-copilot project with an approved publisher prefix.", "Upload the two synthetic knowledge files and create one manual skill for every advertised operation.", "Bind dedicated approved connection references only after replacing the synthetic evidence route.", "Push the draft, replay every locked persona case, and stop before publication or any external action." ], "local_install_prompt": "You are GitHub Copilot running in Agent mode in VS Code. Install and open the AI BAST Sales Qualification Agent from this deployment recipe: https://raw.githubusercontent.com/microsoft/aibast-agents-library/main/solutions/sales-qualification/deployment.json. Own the setup and do not ask me to run terminal commands. Inspect the operating system; verify the Brainstem health endpoint; if needed, use the official installer from the recipe; fetch and install the portable source; verify the expected tool; and replay the recipe smoke test. Preserve the synthetic-only route for score_leads, bant_analysis, create_outreach, assign_leads, setup_tracking, qualification_report. Do not connect customer systems, perform writes, publish, approve pricing, send outreach, or silently substitute another agent. If authentication requires browser confirmation, ask only for that browser step. Stop and report the exact evidence and any failed step. Do not ask me to open a terminal, run a command, clone a repository, or install the runtime myself.", "copilot_studio_prompt": "You are GitHub Copilot running in Agent mode in VS Code with the Microsoft Copilot Studio plugin installed. Prepare the AI BAST Sales Qualification Agent from https://raw.githubusercontent.com/microsoft/aibast-agents-library/main/solutions/sales-qualification/deployment.json. Ask only for the target Power Platform environment ID and publisher prefix. Verify PAC CLI, authenticate if needed, initialize a cli-copilot project, upload the two synthetic knowledge files, and create one skill for each operation: score_leads, bant_analysis, create_outreach, assign_leads, setup_tracking, qualification_report. Bind only dedicated approved connection references, preserve every evidence and approval boundary, pull before push, and replay every locked persona case. Stop before publish and show the generated files, validation evidence, unresolved connection work, and the exact publish command that would require user approval.", "required_connections": [ "Dynamics 365 Sales", "Approved lead-enrichment connector", "Territory and capacity data", "Microsoft Teams" ], "manual_commands": [ "pac auth create", "pac copilot init --name \"Sales Qualification Agent\" --publisher-prefix --authoring-mode cli-copilot --project-dir \"\" --environment \"\"", "pac connection list --environment \"\"", "pac copilot pull --project-dir \"\"", "pac copilot push --project-dir \"\"", "pac copilot publish --bot \"\" --environment \"\"" ], "acceptance_checks": [ "The package remains under solutions/sales-qualification and the portable source resolves to the expected tool.", "Every advertised operation has one uploadable skill and one locked persona-language capture case.", "The approved one-pager filename and source hash remain pinned in the one-pager map.", "Every exact value is identified as synthetic or as a planning assumption rather than a measured customer outcome.", "No CRM write, outreach, approval, pricing change, forecast change, task, alert, document delivery, or customer communication occurs.", "Publishing remains a separate user-approved step after push and validation." ], "hard_mode": [ "Run pac copilot init and confirm settings.mcs.yml preserves the intended environment, language, and publisher identity.", "Author global instructions that route only the supported operations: score_leads, bant_analysis, create_outreach, assign_leads, setup_tracking, qualification_report.", "Upload both synthetic knowledge files and create each operation skill from the package SKILL.md source.", "Create dedicated connection references for the approved production seams; do not reuse ambiguous existing connections.", "Validate schema, instructions, skills, knowledge, and connection bindings; pull before push and replay every locked case.", "Keep the agent in Draft and require explicit confirmation before publish or any external action." ] }, "curated_copy": true, "source": "catalog", "origin": null, "onepager": "21. Sales Qualification Agent one-pager.pptx", "demo_video": null, "has_onepager": true, "has_demo_video": false, "promise_coverage": 60, "is_primary": true, "aliases": [], "package": { "slug": "sales-qualification", "quest_url": "solutions/sales-qualification/quest.html", "manual_tutorial_url": "solutions/sales-qualification/manual-tutorial.html", "export_manifest_url": "solutions/sales-qualification/export-manifest.json" } }, "_demo": { "slug": "sales-qualification", "case_count": 6, "personas": [ "Account Executive", "Business Development Rep.", "Sales Manager" ], "prompts": [ "Which bundled synthetic leads should my team review first, and why did they score that way?", "Show the BANT evidence and missing qualification details for the strongest synthetic leads.", "Draft outreach ideas for the synthetic hot leads, but do not send or schedule any communication." ] }, "_catalog_kind": "solution", "_readiness": [ "Demo proven", "Conversation tested", "Synthetic data", "Production pattern" ] }, { "schema": "rapp-agent/1.0", "name": "@aibast-agents-library/win-loss-analysis", "version": "1.0.0", "display_name": "Win Loss Analysis Agent", "description": "Automates competitive deal analysis to uncover root causes, improve win rates, and guide strategic sales enablement.", "author": "AIBAST", "tags": [ "b2b", "sales", "win-loss", "competitive-intel", "revenue-recovery" ], "category": "b2b_sales", "quality_tier": "verified", "requires_env": [], "dependencies": [ "@rapp/basic-agent" ], "_file": "agents/@aibast-agents-library/b2b_sales_stacks/win_loss_analysis_stack/win_loss_analysis_agent.py", "_size_kb": 81.5, "_lines": 829, "_sha256": "17ecf36b1f6ac899a979d68a410ea630cd6a49336e7378c058c46bff3438bc5f", "_install_prefix": "win_loss_analysis__", "_install_filename": "win_loss_analysis__17ecf36b1f6a_agent.py", "_stack": "win_loss_analysis", "_stack_vertical": "b2b_sales", "_synthetic_data": true, "_added_at": "2026-03-18T13:23:12-04:00", "_solution": { "advertised_name": "Win/Loss Analysis Agent", "sharepoint_list_name": "Win-Loss Analysis Agent", "slot": 22, "executive_summary": "Automates competitive deal analysis to uncover root causes, improve win rates, and guide strategic sales enablement.", "industries": [], "personas": [], "featured_tools": [ "Dynamics 365", "Microsoft Teams" ], "agent_requirements": [ "Microsoft Copilot Studio" ], "capabilities": [ "Aggregate and analyze historical deal data for trend identification", "Synthesize buyer feedback and competitor patterns into actionable insights", "Develop targeted counter-strategies and enablement playbooks for reps", "Model revenue recovery scenarios for leadership decision-making" ], "outcomes": [ "Identified top loss drivers and prioritized addressable gaps", "Equipped sales teams with data-driven objection handling and competitive talk tracks", "Reversed declining win rates through targeted interventions", "Delivered board-ready insights and recovery plans in hours, not weeks" ], "customer_scenario": [ "Loss reasons were anecdotal, making root cause analysis unreliable", "Competitive wins decreased without clear visibility into gaps", "Sales teams had no structured guidance to counter objections effectively", "Leadership lacked actionable insights for board reporting" ], "slide_title": "Win/Loss Analysis Agent", "slide_summary": "Automates competitive deal analysis to uncover root causes, improve win rates, and guide strategic sales enablement.", "scenario_name": "A technology services company", "challenge_intro": "A technology services company faced a sharp decline in enterprise win rates:", "actions_intro": "The tech services company leveraged the Win/Loss Analysis Agent to:", "outcomes_intro": "With the agent\u2019s help, the company:", "opportunity_statements": [ "Perform pattern recognition and trend analysis to identify actionable insights and reverse declining performance", "Analyze buyer feedback and loss reasons to prioritize addressable gaps and strengthen competitive positioning", "Generate targeted strategies and enablement content to equip teams for faster recovery and sustained growth" ], "module_name": null, "module_description": null, "source_verified": true, "sales_headline": "Turn closed-deal evidence into a strategy review", "card_pitch": "Help sales operations, enablement, and leadership compare win and loss patterns, isolate addressable drivers, draft counter-strategies, and evaluate clearly labeled scenarios.", "why_try": "A leader can trace a strategic recommendation back to structured deal evidence instead of relying on anecdotes or treating a model as a commitment.", "customer_challenge_copy": "Loss reasons and buyer feedback are often inconsistent, leaving leaders without a shared basis for enablement, product, or investment decisions.", "microsoft_ai_story": "Microsoft Copilot Studio provides the analysis and strategy-review experience. Dynamics 365 Sales contributes closed-opportunity evidence, approved conversation-intelligence sources contribute buyer feedback, Power BI can visualize trends, and Microsoft Teams supports enablement review.", "business_value": [ "Creates a consistent view of trends, competitors, and loss drivers.", "Connects buyer-feedback themes to draft enablement responses.", "Supports leadership scenario review without claiming realized performance." ], "search_terms": [ "win loss analysis", "closed deal analysis", "loss reasons", "buyer feedback", "competitive intelligence", "sales enablement", "revenue scenario", "Dynamics 365 Sales", "conversation intelligence", "Power BI sales", "Microsoft Teams enablement", "Copilot Studio" ], "journey_stage": "Learn", "blueprint_role": "Creates a governed feedback loop from closed-deal evidence to enablement and leadership planning.", "sample_prompts": [ { "label": "Compare win and loss patterns", "prompt": "Compare the bundled synthetic Q3 and Q2 win and loss patterns and show where the decline is concentrated.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=win-loss-analysis-win-loss-overview" }, { "label": "Find the root causes", "prompt": "Identify the evidence-backed synthetic loss drivers and buyer feedback themes that enablement should review.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=win-loss-analysis-root-cause-analysis" }, { "label": "Draft counter-strategies", "prompt": "Draft counter-strategy and talk-track options from the synthetic loss evidence for enablement review.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=win-loss-analysis-counter-strategies" }, { "label": "Model value scenarios", "prompt": "Model synthetic intervention scenarios without presenting them as realized or committed revenue.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=win-loss-analysis-revenue-impact" }, { "label": "Draft the board narrative", "prompt": "Draft a board-level synthetic win and loss narrative with all investment and performance values labeled as scenarios.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=win-loss-analysis-board-presentation" }, { "label": "Summarize next steps", "prompt": "Summarize the synthetic findings and candidate next steps without activating programs or approvals.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=win-loss-analysis-action-summary" } ], "architecture": { "business_flow": [ "Sales operations, enablement, or sales leadership", "Copilot experience", "Win/Loss Analysis Agent", "Synthetic closed-deal and buyer-feedback evidence", "Human strategy and investment review" ], "capabilities": [ { "operation": "win_loss_overview", "name": "Win Loss Overview", "purpose": "Compares structured outcome and competitor patterns." }, { "operation": "root_cause_analysis", "name": "Root Cause Analysis", "purpose": "Ranks evidence-backed loss drivers and buyer-feedback themes." }, { "operation": "counter_strategies", "name": "Counter Strategies", "purpose": "Drafts near-term and longer-term enablement options." }, { "operation": "revenue_impact", "name": "Revenue Impact", "purpose": "Models clearly labeled value scenarios rather than revenue commitments." }, { "operation": "board_presentation", "name": "Board Presentation", "purpose": "Drafts a leadership narrative with explicit scenario boundaries." }, { "operation": "action_summary", "name": "Action Summary", "purpose": "Summarizes findings and candidate next steps without activation." } ], "easy_mode": [ "Verify Power Platform CLI and authenticate to the target environment.", "Initialize a modern cli-copilot project with an approved publisher prefix.", "Upload the two synthetic knowledge files and create one manual skill for every advertised operation.", "Bind dedicated approved connection references only after replacing the synthetic evidence route.", "Push the draft, replay every locked persona case, and stop before publication or any external action." ], "local_install_prompt": "You are GitHub Copilot running in Agent mode in VS Code. Install and open the AI BAST Win/Loss Analysis Agent from this deployment recipe: https://raw.githubusercontent.com/microsoft/aibast-agents-library/main/solutions/win-loss-analysis/deployment.json. Own the setup and do not ask me to run terminal commands. Inspect the operating system; verify the Brainstem health endpoint; if needed, use the official installer from the recipe; fetch and install the portable source; verify the expected tool; and replay the recipe smoke test. Preserve the synthetic-only route for win_loss_overview, root_cause_analysis, counter_strategies, revenue_impact, board_presentation, action_summary. Do not connect customer systems, perform writes, publish, approve pricing, send outreach, or silently substitute another agent. If authentication requires browser confirmation, ask only for that browser step. Stop and report the exact evidence and any failed step. Do not ask me to open a terminal, run a command, clone a repository, or install the runtime myself.", "copilot_studio_prompt": "You are GitHub Copilot running in Agent mode in VS Code with the Microsoft Copilot Studio plugin installed. Prepare the AI BAST Win/Loss Analysis Agent from https://raw.githubusercontent.com/microsoft/aibast-agents-library/main/solutions/win-loss-analysis/deployment.json. Ask only for the target Power Platform environment ID and publisher prefix. Verify PAC CLI, authenticate if needed, initialize a cli-copilot project, upload the two synthetic knowledge files, and create one skill for each operation: win_loss_overview, root_cause_analysis, counter_strategies, revenue_impact, board_presentation, action_summary. Bind only dedicated approved connection references, preserve every evidence and approval boundary, pull before push, and replay every locked persona case. Stop before publish and show the generated files, validation evidence, unresolved connection work, and the exact publish command that would require user approval.", "required_connections": [ "Dynamics 365 Sales", "Approved buyer-feedback or conversation-intelligence source", "Competitive intelligence source", "Power BI", "Microsoft Teams" ], "manual_commands": [ "pac auth create", "pac copilot init --name \"Win/Loss Analysis Agent\" --publisher-prefix --authoring-mode cli-copilot --project-dir \"\" --environment \"\"", "pac connection list --environment \"\"", "pac copilot pull --project-dir \"\"", "pac copilot push --project-dir \"\"", "pac copilot publish --bot \"\" --environment \"\"" ], "acceptance_checks": [ "The package remains under solutions/win-loss-analysis and the portable source resolves to the expected tool.", "Every advertised operation has one uploadable skill and one locked persona-language capture case.", "The approved one-pager filename and source hash remain pinned in the one-pager map.", "Every exact value is identified as synthetic or as a planning assumption rather than a measured customer outcome.", "No CRM write, outreach, approval, pricing change, forecast change, task, alert, document delivery, or customer communication occurs.", "Publishing remains a separate user-approved step after push and validation." ], "hard_mode": [ "Run pac copilot init and confirm settings.mcs.yml preserves the intended environment, language, and publisher identity.", "Author global instructions that route only the supported operations: win_loss_overview, root_cause_analysis, counter_strategies, revenue_impact, board_presentation, action_summary.", "Upload both synthetic knowledge files and create each operation skill from the package SKILL.md source.", "Create dedicated connection references for the approved production seams; do not reuse ambiguous existing connections.", "Validate schema, instructions, skills, knowledge, and connection bindings; pull before push and replay every locked case.", "Keep the agent in Draft and require explicit confirmation before publish or any external action." ] }, "curated_copy": true, "source": "catalog", "origin": null, "onepager": "22. Win Loss Analysis Agent one-pager.pptx", "demo_video": null, "has_onepager": true, "has_demo_video": false, "promise_coverage": 70, "is_primary": true, "aliases": [], "package": { "slug": "win-loss-analysis", "quest_url": "solutions/win-loss-analysis/quest.html", "manual_tutorial_url": "solutions/win-loss-analysis/manual-tutorial.html", "export_manifest_url": "solutions/win-loss-analysis/export-manifest.json" } }, "_demo": { "slug": "win-loss-analysis", "case_count": 6, "personas": [ "Enablement Manager", "Sales Leader", "Sales Operations Manager" ], "prompts": [ "Compare the bundled synthetic Q3 and Q2 win and loss patterns and show where the decline is concentrated.", "Identify the evidence-backed synthetic loss drivers and buyer feedback themes that enablement should review.", "Draft counter-strategy and talk-track options from the synthetic loss evidence for enablement review." ] }, "_catalog_kind": "solution", "_readiness": [ "Demo proven", "Conversation tested", "Synthetic data", "Production pattern" ] }, { "schema": "rapp-agent/1.0", "name": "@aibast-agents-library/cart-abandonment-recovery", "version": "1.0.0", "display_name": "Cart Abandonment Recovery Agent", "description": "Draft privacy-safe abandoned-cart analysis, recovery concepts, incentive scenarios, and aggregate conversion reporting for human review.", "author": "AIBAST", "tags": [ "cart-abandonment", "recovery", "ecommerce", "conversion", "email", "b2c" ], "category": "b2c_sales", "quality_tier": "verified", "requires_env": [], "dependencies": [ "@rapp/basic-agent" ], "_file": "agents/@aibast-agents-library/b2c_sales_stacks/cart_abandonment_recovery_stack/cart_abandonment_recovery_agent.py", "_size_kb": 21.4, "_lines": 443, "_sha256": "e521927e4c121e38115e630ab013370dab9964f8c6b8672daac01a1afb1cc35b", "_install_prefix": "cart_abandonment_recovery__", "_install_filename": "cart_abandonment_recovery__e521927e4c12_agent.py", "_stack": "cart_abandonment_recovery", "_stack_vertical": "b2c_sales", "_synthetic_data": true, "_added_at": "2026-03-18T13:23:12-04:00", "_solution": { "advertised_name": "Cart Abandonment Recovery Agent", "sharepoint_list_name": "Abandoned Cart Recovery Agent", "slot": 22, "executive_summary": "Automate abandoned cart analysis and recovery campaigns to convert lost sales, protect margins, and improve customer engagement.", "industries": [ "Cross-Industry", "Software Tech" ], "personas": [ "Marketing Manager", "Digital Marketing Lead", "Growth Manager" ], "featured_tools": [ "Dynamics 365", "Microsoft Teams", "Outlook" ], "agent_requirements": [ "Microsoft Copilot Studio" ], "capabilities": [ "Segment abandoned carts by value and recovery likelihood", "Generate personalized offers and urgency messaging for each segment", "Deploy multi-touch campaigns across email, SMS, and retargeting with optimized timing", "Forecast recovery rates and revenue impact for leadership reporting" ], "outcomes": [ "Increased recovery rates through precision targeting and personalized incentives", "Protected margins with discount guardrails and alternative offers", "Reduced manual workloads and response times", "Improved decision-making with real-time performance dashboards" ], "customer_scenario": [ "Manual recovery campaigns slowed engagement and missed conversion windows", "Generic discounting eroded margins without improving recovery rates", "Lack of segmentation led to poor targeting and wasted incentives", "Leadership lacked full visibility into recovery performance and ROI" ], "slide_title": "Cart Abandonment Recovery Agent", "slide_summary": "Automates abandoned cart analysis and recovery campaigns to convert lost sales, protect margins, and improve customer engagement.", "scenario_name": "An eCommerce company", "challenge_intro": "An eCommerce company faced high cart abandonment rates during peak season:", "actions_intro": "The eCommerce company leveraged the Cart Abandonment Recovery Agent to:", "outcomes_intro": "With the agent\u2019s help, the company:", "opportunity_statements": [ "Monitor and analyze abandoned carts in real time to recover lost sales", "Generate personalized offers to maximize conversion and maintain profitability", "Orchestrate recovery campaigns to improve engagement and recovery rates", "Forecast recovery and revenue impact to optimize strategy and campaign ROI" ], "module_name": null, "module_description": null, "source_verified": true, "sales_headline": "Turn cart friction into consent-aware recovery planning without touching a shopper or cart.", "card_pitch": "Analyzes anonymous synthetic cart patterns, drafts neutral channel sequences, compares margin-aware incentive concepts, and reviews aggregate conversion scenarios.", "why_try": "It helps growth and digital teams pressure-test recovery ideas while preserving consent, margin governance, and activation controls.", "customer_challenge_copy": "Recovery programs can become slow, generic, over-discounted, and difficult to evaluate when journey friction and margin tradeoffs are unclear.", "microsoft_ai_story": "A portable planning agent converts anonymous commerce-style records into review briefs that marketing teams can refine before approved systems handle any contact or offer.", "business_value": [ "Clearer diagnosis of cart friction", "More disciplined review of channel and incentive options", "Stronger consent and activation governance" ], "search_terms": [ "cart abandonment", "recovery planning", "ecommerce conversion", "incentive scenarios", "consent" ], "journey_stage": "Recover and learn", "blueprint_role": "Anonymous recovery planning", "sample_prompts": [ { "label": "Diagnose anonymous cart friction", "prompt": "Show where the anonymous synthetic cart journey is breaking down and what needs review.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=cart-abandonment-recovery-abandonment-analysis" }, { "label": "Draft a consent-aware sequence", "prompt": "Outline a consent-aware draft sequence without sending or scheduling anything.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=cart-abandonment-recovery-recovery-campaign" }, { "label": "Compare incentive concepts", "prompt": "Compare margin-aware value scenarios and keep every concept behind approval.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=cart-abandonment-recovery-incentive-optimization" }, { "label": "Review recovery scenarios", "prompt": "Summarize the fixed recovery metrics and separate benchmarks from planning estimates.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=cart-abandonment-recovery-conversion-tracking" } ], "architecture": { "business_flow": [ "Marketing Manager, Digital Marketing Lead, and Growth Manager", "Microsoft 365 Copilot or Copilot Studio", "Cart Abandonment Recovery Agent", "Approved aggregate commerce analytics", "Consent and campaign review workflow", "Authorized activation systems" ], "capabilities": [ { "operation": "abandonment_analysis", "name": "Anonymous abandonment analysis", "purpose": "Finds aggregate exit and value patterns without identifying or contacting a shopper." }, { "operation": "recovery_campaign", "name": "Consent-aware campaign draft", "purpose": "Outlines review-only channel timing and neutral content concepts." }, { "operation": "incentive_optimization", "name": "Margin-aware incentive scenarios", "purpose": "Compares optional value concepts without issuing an offer." }, { "operation": "conversion_tracking", "name": "Synthetic conversion review", "purpose": "Summarizes fixed recovery metrics and planning assumptions." } ], "easy_mode": [ "Initialize a modern Copilot Studio project in the target environment.", "Provide this blueprint and the deployment recipe to the Copilot Studio Architect workflow.", "Upload the packaged global instructions, synthetic knowledge files, and operation skills.", "Bind only approved read-only connections and keep all write actions separate.", "Push the draft, replay every locked prompt, and stop before publishing." ], "local_install_prompt": "You are GitHub Copilot running in Agent mode in VS Code. Install and open the AI BAST Cart Abandonment Recovery Agent from https://raw.githubusercontent.com/microsoft/aibast-agents-library/main/solutions/cart-abandonment-recovery/deployment.json. Own the setup, verify the source path, confirm the expected tool in Brainstem health, run the smoke prompt, and open the local UI. Use synthetic data only, preserve every safety boundary, and do not publish or silently substitute another agent. Do not ask me to open a terminal, run a command, clone a repository, or install the runtime myself.", "copilot_studio_prompt": "You are GitHub Copilot running in Agent mode with the Microsoft Copilot Studio plugin. Prepare a draft Cart Abandonment Recovery Agent from its deployment recipe. Ask only for the target environment and publisher prefix. Upload the packaged global instructions, two knowledge files, and one skill per operation; bind approved read-only connections; validate all locked prompts; and stop before publish with every external action and authorization gate visible. Stop before publish.", "required_connections": [ "Commerce analytics through an approved aggregate read-only feed", "Outlook or Teams only after separate approval and implementation" ], "manual_commands": [ "pac auth create", "pac copilot init --name \"Cart Abandonment Recovery Agent\" --publisher-prefix --authoring-mode cli-copilot --project-dir \"\" --environment \"\"", "pac connection list --environment \"\"", "pac copilot pull --project-dir \"\"", "pac copilot push --project-dir \"\"", "pac copilot publish --bot \"\" --environment \"\"" ], "acceptance_checks": [ "Every advertised capability is represented by a deterministic operation and uploadable skill.", "The exact portable source stem is present in the capture contract.", "Both synthetic knowledge files and the global instructions are packaged under the standard solution directory.", "Outputs preserve privacy, recommendation-only behavior, and explicit human approval gates.", "All locked persona prompts route to deterministic synthetic evidence.", "Publishing remains a separate authorized step." ], "hard_mode": [ "Create the agent manually in Copilot Studio.", "Enter the packaged global instructions.", "Upload both synthetic knowledge files.", "Create one skill for each operation.", "Bind only approved read-only connections.", "Replay every locked case and verify safety gates before publishing." ] }, "curated_copy": true, "source": "catalog", "origin": null, "onepager": "25. Cart Abandonment Recovery Agent one-pager.pptx", "demo_video": "25-Cart Abandonment Recovery Agent.mp4", "has_onepager": true, "has_demo_video": true, "promise_coverage": 60, "is_primary": true, "aliases": [], "package": { "slug": "cart-abandonment-recovery", "quest_url": "solutions/cart-abandonment-recovery/quest.html", "manual_tutorial_url": "solutions/cart-abandonment-recovery/manual-tutorial.html", "export_manifest_url": "solutions/cart-abandonment-recovery/export-manifest.json" } }, "_demo": { "slug": "cart-abandonment-recovery", "case_count": 7, "personas": [ "Digital Marketing Lead", "Growth Manager", "Marketing Manager" ], "prompts": [ "As Marketing Manager, show where the anonymous synthetic cart journey is breaking down and what needs review.", "As Digital Marketing Lead, outline a consent-aware draft sequence without sending or scheduling anything.", "As Growth Manager, compare margin-aware value scenarios and keep every concept behind approval." ] }, "_catalog_kind": "solution", "_readiness": [ "Demo proven", "Conversation tested", "Synthetic data", "Production pattern" ] }, { "schema": "rapp-agent/1.0", "name": "@aibast-agents-library/customer-loyalty-rewards", "version": "1.1.0", "display_name": "Customer Loyalty and Rewards Agent", "description": "Provide synthetic loyalty insights: churn-risk segments, top at-risk members, win-back offer drafts, launch-ready campaign plans with projected results, program improvements, points summaries, reward options, and tier analysis, without contacting members or issuing or redeeming benefits.", "author": "AIBAST", "tags": [ "loyalty", "rewards", "points", "retention", "churn", "win-back", "tier", "b2c" ], "category": "b2c_sales", "quality_tier": "verified", "requires_env": [], "dependencies": [ "@rapp/basic-agent" ], "_file": "agents/@aibast-agents-library/b2c_sales_stacks/customer_loyalty_rewards_stack/customer_loyalty_rewards_agent.py", "_size_kb": 30.7, "_lines": 637, "_sha256": "b09a5236086a89fbfbfe6b98c1f89f10da7ce5468e3c852a1b56011ad9d7f658", "_install_prefix": "customer_loyalty_rewards__", "_install_filename": "customer_loyalty_rewards__b09a5236086a_agent.py", "_stack": "customer_loyalty_rewards", "_stack_vertical": "b2c_sales", "_synthetic_data": true, "_added_at": "2026-03-18T13:23:12-04:00", "_solution": { "advertised_name": "Customer Loyalty and Rewards Agent", "sharepoint_list_name": "Loyalty Insights Agent", "slot": 43, "executive_summary": "Deliver AI-driven loyalty insights and planning to reduce points liability, improve engagement results, and boost member retention.", "industries": [ "Cross-Industry" ], "personas": [ "Loyalty Program Director", "CRM Manager", "Marketing Leader" ], "featured_tools": [ "Dynamics 365", "Outlook", "Microsoft Teams" ], "agent_requirements": [ "Microsoft Copilot Studio" ], "capabilities": [ "Segment members by engagement level, churn risk, and point expiration", "Tailor offers and incentives based on purchase behavior and preferences", "Deliver targeted outreach through the highest\u2011engagement channels", "Recommend structural program improvements to reduce liability and improve redemption" ], "outcomes": [ "Re\u2011engaged at\u2011risk members with timely, personalized campaigns", "Reduced point liability by prompting relevant and high\u2011value redemptions", "Improved program ROI with precise targeting and optimized reward strategies", "Strengthened loyalty with offers aligned to individual intent" ], "customer_scenario": [ "High\u2011value members were going inactive without early warning", "Points accumulated faster than they were redeemed", "Campaigns were broad and generic, delivering low response rates", "Marketing teams lacked insight into churn signals and offer effectiveness" ], "slide_title": "Customer Loyalty and Rewards Agent", "slide_summary": "Delivers AI\u2011driven loyalty insights and planning to reduce points liability, improve engagement results, and boost member retention.", "scenario_name": "A large brick-and-mortar retailer", "challenge_intro": "A large retail brand faced declining member engagement and rising points liability.", "actions_intro": "The retailer used the Customer Loyalty and Rewards Agent to:", "outcomes_intro": "With the agent, the retailer:", "opportunity_statements": [ "Analyze loyalty program health and member behavior to identify churn risk and engagement drivers", "Generate personalized offers and campaigns that improve response and engagement rates", "Optimize loyalty program structure and rewards to improve redemption performance and manage point liability" ], "module_name": null, "module_description": null, "source_verified": true, "sales_headline": "Understand loyalty health and reward design without changing a member account.", "card_pitch": "Provides synthetic program health, informational points views, review-only reward options, and tier-structure analysis.", "why_try": "It helps loyalty, CRM, and marketing leaders examine program tradeoffs while keeping contact, eligibility, balances, benefits, and redemptions in authorized systems.", "customer_challenge_copy": "Program teams need a shared view of engagement, point liability, reward relevance, and tier design without turning analysis into an account action.", "microsoft_ai_story": "The portable agent turns CRM- and ERP-style synthetic loyalty records into governed planning briefs for Microsoft collaboration surfaces.", "business_value": [ "Clearer program-health discussions", "More transparent reward and tier tradeoffs", "Stronger separation between insight and benefit issuance" ], "search_terms": [ "customer loyalty", "rewards", "points", "tiers", "retention planning" ], "journey_stage": "Retain and optimize", "blueprint_role": "Informational loyalty planning", "sample_prompts": [ { "label": "Review loyalty program health", "prompt": "Summarize the synthetic program health and governance boundary.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=customer-loyalty-rewards-loyalty-dashboard" }, { "label": "Explain an informational balance", "prompt": "Explain the synthetic Gold member balance without changing the ledger.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=customer-loyalty-rewards-points-summary" }, { "label": "Compare reward options", "prompt": "Show review-only reward options without creating offers or issuing benefits.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=customer-loyalty-rewards-reward-recommendations" }, { "label": "Analyze tier structure", "prompt": "Review the synthetic tier structure without changing member status.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=customer-loyalty-rewards-tier-analysis" } ], "architecture": { "business_flow": [ "Loyalty Program Director, CRM Manager, and Marketing Leader", "Microsoft 365 Copilot or Copilot Studio", "Customer Loyalty and Rewards Agent", "Approved read-only loyalty and CRM data", "Microsoft Teams program review", "Authorized rewards workflow" ], "capabilities": [ { "operation": "loyalty_dashboard", "name": "Synthetic program health", "purpose": "Summarizes aggregate loyalty evidence without changing the program." }, { "operation": "points_summary", "name": "Informational points summary", "purpose": "Explains a synthetic balance without changing a ledger." }, { "operation": "reward_recommendations", "name": "Review-only reward options", "purpose": "Compares preference-matched concepts without creating an offer or issuing a benefit." }, { "operation": "tier_analysis", "name": "Tier structure analysis", "purpose": "Reviews thresholds and progress without changing member status." } ], "easy_mode": [ "Initialize a modern Copilot Studio project in the target environment.", "Provide this blueprint and the deployment recipe to the Copilot Studio Architect workflow.", "Upload the packaged global instructions, synthetic knowledge files, and operation skills.", "Bind only approved read-only connections and keep all write actions separate.", "Push the draft, replay every locked prompt, and stop before publishing." ], "local_install_prompt": "You are GitHub Copilot running in Agent mode in VS Code. Install and open the AI BAST Customer Loyalty and Rewards Agent from https://raw.githubusercontent.com/microsoft/aibast-agents-library/main/solutions/customer-loyalty-rewards/deployment.json. Own the setup, verify the source path, confirm the expected tool in Brainstem health, run the smoke prompt, and open the local UI. Use synthetic data only, preserve every safety boundary, and do not publish or silently substitute another agent. Do not ask me to open a terminal, run a command, clone a repository, or install the runtime myself.", "copilot_studio_prompt": "You are GitHub Copilot running in Agent mode with the Microsoft Copilot Studio plugin. Prepare a draft Customer Loyalty and Rewards Agent from its deployment recipe. Ask only for the target environment and publisher prefix. Upload the packaged global instructions, two knowledge files, and one skill per operation; bind approved read-only connections; validate all locked prompts; and stop before publish with every external action and authorization gate visible. Stop before publish.", "required_connections": [ "Dynamics CRM and ERP through approved read-only integrations", "Microsoft Teams for program review" ], "manual_commands": [ "pac auth create", "pac copilot init --name \"Customer Loyalty and Rewards Agent\" --publisher-prefix --authoring-mode cli-copilot --project-dir \"\" --environment \"\"", "pac connection list --environment \"\"", "pac copilot pull --project-dir \"\"", "pac copilot push --project-dir \"\"", "pac copilot publish --bot \"\" --environment \"\"" ], "acceptance_checks": [ "Every advertised capability is represented by a deterministic operation and uploadable skill.", "The exact portable source stem is present in the capture contract.", "Both synthetic knowledge files and the global instructions are packaged under the standard solution directory.", "Outputs preserve privacy, recommendation-only behavior, and explicit human approval gates.", "All locked persona prompts route to deterministic synthetic evidence.", "Publishing remains a separate authorized step." ], "hard_mode": [ "Create the agent manually in Copilot Studio.", "Enter the packaged global instructions.", "Upload both synthetic knowledge files.", "Create one skill for each operation.", "Bind only approved read-only connections.", "Replay every locked case and verify safety gates before publishing." ] }, "curated_copy": true, "source": "catalog", "origin": null, "onepager": "45. Customer Loyalty and Rewards Agent one-pager.pptx", "demo_video": "45-Customer Loyalty Rewards Agent.mp4", "has_onepager": true, "has_demo_video": true, "promise_coverage": 35, "is_primary": true, "aliases": [], "package": { "slug": "customer-loyalty-rewards", "quest_url": "solutions/customer-loyalty-rewards/quest.html", "manual_tutorial_url": "solutions/customer-loyalty-rewards/manual-tutorial.html", "export_manifest_url": "solutions/customer-loyalty-rewards/export-manifest.json" } }, "_demo": { "slug": "customer-loyalty-rewards", "case_count": 10, "personas": [ "CRM Manager", "Loyalty Program Director", "Marketing Leader" ], "prompts": [ "As Loyalty Program Director, summarize the synthetic program health and governance boundary.", "As CRM Manager, explain the synthetic Gold member balance without changing the ledger.", "As Marketing Leader, show review-only reward options without creating offers or issuing benefits." ] }, "_catalog_kind": "solution", "_readiness": [ "Demo proven", "Conversation tested", "Synthetic data", "Production pattern" ] }, { "schema": "rapp-agent/1.0", "name": "@aibast-agents-library/omnichannel-engagement", "version": "1.0.0", "display_name": "Omnichannel Engagement Agent", "description": "Unify one customer's interactions across channels so service and sales teams pick up where the customer left off: journey, unresolved questions, best channel and timing, proactive engagement drafts, and handoff context; plus aggregate channel analytics.", "author": "AIBAST", "tags": [ "omnichannel", "engagement", "journey", "attribution", "campaign", "b2c" ], "category": "b2c_sales", "quality_tier": "verified", "requires_env": [], "dependencies": [ "@rapp/basic-agent" ], "_file": "agents/@aibast-agents-library/b2c_sales_stacks/omnichannel_engagement_stack/omnichannel_engagement_agent.py", "_size_kb": 27.6, "_lines": 582, "_sha256": "f29197f964d4fbdf2c83b93a07bcb4df11bc056ae80bc79c9daf555999843621", "_install_prefix": "omnichannel_engagement__", "_install_filename": "omnichannel_engagement__f29197f964d4_agent.py", "_stack": "omnichannel_engagement", "_stack_vertical": "b2c_sales", "_synthetic_data": true, "_added_at": "2026-03-18T13:23:12-04:00", "_solution": { "advertised_name": "Omnichannel Engagement Agent", "sharepoint_list_name": "Cross-Channel Engagement Agent", "slot": 27, "executive_summary": "Deliver a single, unified view of cross-channel interactions for more strategic, streamlined support and stronger engagement.", "industries": [ "Cross-Industry", "Retail" ], "personas": [ "Customer Experience Leader", "Digital Engagement Manager", "Contact Center Supervisor" ], "featured_tools": [ "Dynamics 365", "Outlook" ], "agent_requirements": [ "Microsoft Copilot Studio" ], "capabilities": [ "Consolidate interactions into a single, continuous customer view", "Surface unresolved questions and journey friction instantly", "Recommend the best channel and message timing for each customer", "Strengthen continuity across touchpoints to deliver a consistent, connected customer experience" ], "outcomes": [ "Resolved issues in a single interaction with complete context", "Increased conversion by delivering the right content at key moments", "Reduced repeat contacts across channels", "Improved customer satisfaction through consistent, personalized engagement" ], "customer_scenario": [ "Conversations spanned across app, chat, and phone, forcing customers to repeat themselves", "High\u2011intent shoppers dropped off when switching channels", "Engagement campaigns were sent through the wrong channels, lowering response rates" ], "slide_title": "Omnichannel Engagement Agent", "slide_summary": "Delivers a single, unified view of cross-channel interactions for more strategic, streamlined support and stronger engagement.", "scenario_name": "A national apparel retailer", "challenge_intro": "A national apparel retailer saw inconsistent customer experiences across digital and support channels:", "actions_intro": "The retailer used the Omnichannel Engagement Agent to:", "outcomes_intro": "With the Agent, the retailer:", "opportunity_statements": [ "Aggregate customer interactions into a unified context to speed up resolution", "Map customer journeys and detect friction points to quickly resolve blockers", "Orchestrate channel\u2011optimized engagement to increase response rates and drive higher conversion" ], "module_name": null, "module_description": null, "source_verified": true, "sales_headline": "See cross-channel friction clearly without building a person-level surveillance layer.", "card_pitch": "Compares aggregate channels, maps journey archetypes, drafts consent-aware engagement experiments, and reviews synthetic attribution evidence.", "why_try": "It gives experience, digital, and contact-center leaders a shared picture of continuity and friction while avoiding identity stitching and outreach.", "customer_challenge_copy": "Channel teams can optimize locally while customers experience broken handoffs, repeated context, and inconsistent engagement.", "microsoft_ai_story": "The agent turns aggregate CRM, commerce, and contact-center style measures into a common review narrative for Microsoft collaboration surfaces.", "business_value": [ "A more coherent view of journey friction", "Better-governed channel experiments", "Clearer attribution assumptions and limitations" ], "search_terms": [ "omnichannel", "customer journey", "contact center", "channel performance", "attribution" ], "journey_stage": "Understand and optimize", "blueprint_role": "Aggregate journey intelligence", "sample_prompts": [ { "label": "Compare aggregate channels", "prompt": "As Customer Experience Leader, compare the aggregate channel picture and call out continuity risks.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=omnichannel-engagement-channel-performance" }, { "label": "Find journey friction", "prompt": "As Contact Center Supervisor, identify handoff friction across the aggregate journey archetypes.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=omnichannel-engagement-journey-analysis" }, { "label": "Draft consent-aware experiments", "prompt": "As Digital Engagement Manager, propose consent-aware experiments with frequency caps and no outreach.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=omnichannel-engagement-engagement-optimization" }, { "label": "Review attribution evidence", "prompt": "As Digital Engagement Manager, summarize the synthetic attribution view and avoid causal claims.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=omnichannel-engagement-campaign-attribution" }, { "label": "Contact Center Supervisor: Customer Journey", "prompt": "Show me the customer's journey across all channels and help me pick up where they left off.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=omnichannel-engagement-customer-journey" }, { "label": "Contact Center Supervisor: Unresolved Issues", "prompt": "Show me what Sarah has been asking about and where we dropped the ball.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=omnichannel-engagement-unresolved-issues" }, { "label": "Digital Engagement Manager: Channel Recommendation", "prompt": "Which channel and timing should we use to reach Sarah for this issue and in the future?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=omnichannel-engagement-channel-recommendation" }, { "label": "Digital Engagement Manager: Proactive Plan", "prompt": "What proactive engagement should we plan for Sarah after this purchase, or if she doesn't convert?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=omnichannel-engagement-proactive-plan" }, { "label": "Customer Experience Leader: Handoff Package", "prompt": "Prepare the context in case Sarah needs to be handed off to another agent.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=omnichannel-engagement-handoff-package" } ], "architecture": { "business_flow": [ "Customer Experience Leader, Digital Engagement Manager, and Contact Center Supervisor", "Microsoft 365 Copilot or Copilot Studio", "Omnichannel Engagement Agent", "Approved aggregate CRM, commerce, and contact-center analytics", "Governed review workflow", "Authorized engagement systems" ], "capabilities": [ { "operation": "channel_performance", "name": "Aggregate channel performance", "purpose": "Compares channel measures without identity stitching." }, { "operation": "journey_analysis", "name": "Privacy-safe journey analysis", "purpose": "Uses aggregate archetypes to identify handoff friction." }, { "operation": "engagement_optimization", "name": "Consent-aware optimization", "purpose": "Drafts controlled experiments with consent and frequency caps." }, { "operation": "campaign_attribution", "name": "Synthetic attribution review", "purpose": "Explains fixed attribution evidence without claiming causality." }, { "operation": "customer_journey", "name": "Customer Journey", "purpose": "Customer Journey" }, { "operation": "unresolved_issues", "name": "Unresolved Issues", "purpose": "Unresolved Issues" }, { "operation": "channel_recommendation", "name": "Channel Recommendation", "purpose": "Channel Recommendation" }, { "operation": "proactive_plan", "name": "Proactive Plan", "purpose": "Proactive Plan" }, { "operation": "handoff_package", "name": "Handoff Package", "purpose": "Handoff Package" } ], "easy_mode": [ "Initialize a modern Copilot Studio project in the target environment.", "Provide this blueprint and the deployment recipe to the Copilot Studio Architect workflow.", "Upload the packaged global instructions, synthetic knowledge files, and operation skills.", "Bind only approved read-only connections and keep all write actions separate.", "Push the draft, replay every locked prompt, and stop before publishing." ], "local_install_prompt": "You are GitHub Copilot running in Agent mode in VS Code. Install and open the AI BAST Omnichannel Engagement Agent from https://raw.githubusercontent.com/microsoft/aibast-agents-library/main/solutions/omnichannel-engagement/deployment.json. Own the setup, verify the source path, confirm the expected tool in Brainstem health, run the smoke prompt, and open the local UI. Use synthetic data only, preserve every safety boundary, and do not publish or silently substitute another agent. Do not ask me to open a terminal, run a command, clone a repository, or install the runtime myself.", "copilot_studio_prompt": "You are GitHub Copilot running in Agent mode with the Microsoft Copilot Studio plugin. Prepare a draft Omnichannel Engagement Agent from its deployment recipe. Ask only for the target environment and publisher prefix. Upload the packaged global instructions, two knowledge files, and one skill per operation; bind approved read-only connections; validate all locked prompts; and stop before publish with every external action and authorization gate visible. Stop before publish.", "required_connections": [ "Dynamics CRM and Commerce through approved aggregate integrations", "Contact-center analytics through a governed aggregate feed" ], "manual_commands": [ "pac auth create", "pac copilot init --name \"Omnichannel Engagement Agent\" --publisher-prefix --authoring-mode cli-copilot --project-dir \"\" --environment \"\"", "pac connection list --environment \"\"", "pac copilot pull --project-dir \"\"", "pac copilot push --project-dir \"\"", "pac copilot publish --bot \"\" --environment \"\"" ], "acceptance_checks": [ "Every advertised capability is represented by a deterministic operation and uploadable skill.", "The exact portable source stem is present in the capture contract.", "Both synthetic knowledge files and the global instructions are packaged under the standard solution directory.", "Outputs preserve privacy, recommendation-only behavior, and explicit human approval gates.", "All locked persona prompts route to deterministic synthetic evidence.", "Publishing remains a separate authorized step." ], "hard_mode": [ "Create the agent manually in Copilot Studio.", "Enter the packaged global instructions.", "Upload both synthetic knowledge files.", "Create one skill for each operation.", "Bind only approved read-only connections.", "Replay every locked case and verify safety gates before publishing." ] }, "curated_copy": true, "source": "catalog", "origin": null, "onepager": "30. Omnichannel Engagement Agent one-pager.pptx", "demo_video": "30-Omnichannel Engagement Agent.mp4", "has_onepager": true, "has_demo_video": true, "promise_coverage": 50, "is_primary": true, "aliases": [], "package": { "slug": "omnichannel-engagement", "quest_url": "solutions/omnichannel-engagement/quest.html", "manual_tutorial_url": "solutions/omnichannel-engagement/manual-tutorial.html", "export_manifest_url": "solutions/omnichannel-engagement/export-manifest.json" } }, "_demo": { "slug": "omnichannel-engagement", "case_count": 9, "personas": [ "Contact Center Supervisor", "Customer Experience Leader", "Digital Engagement Manager" ], "prompts": [ "As Customer Experience Leader, compare the aggregate channel picture and call out continuity risks.", "As Contact Center Supervisor, identify handoff friction across the aggregate journey archetypes.", "As Digital Engagement Manager, propose consent-aware experiments with frequency caps and no outreach." ] }, "_catalog_kind": "solution", "_readiness": [ "Demo proven", "Conversation tested", "Synthetic data", "Production pattern" ] }, { "schema": "rapp-agent/1.0", "name": "@aibast-agents-library/personalized-shopping-assistant", "version": "1.1.0", "display_name": "Personalized Shopping Agent", "description": "Draft opt-in product, style, availability, and outfit recommendations using synthetic non-sensitive preferences for human review.", "author": "AIBAST", "tags": [ "shopping", "personalization", "recommendations", "style", "inventory", "b2c" ], "category": "b2c_sales", "quality_tier": "verified", "requires_env": [], "dependencies": [ "@rapp/basic-agent" ], "_file": "agents/@aibast-agents-library/b2c_sales_stacks/personalized_shopping_assistant_stack/personalized_shopping_assistant_agent.py", "_size_kb": 29.8, "_lines": 569, "_sha256": "e4fb09dc4fe8ad2cdacfc5949d65c9e781a515278a8e96fe1d2bd3aa8c4efa19", "_install_prefix": "personalized_shopping_assistant__", "_install_filename": "personalized_shopping_assistant__e4fb09dc4fe8_agent.py", "_stack": "personalized_shopping_assistant", "_stack_vertical": "b2c_sales", "_synthetic_data": true, "_added_at": "2026-03-18T13:23:12-04:00", "_solution": { "advertised_name": "Personalized Shopping Agent", "sharepoint_list_name": "Personal Styling Agent", "slot": 38, "executive_summary": "Deliver intelligent personal styling to strengthen customer experience, increase revenue, and elevate associate efficiency at scale.", "industries": [ "Retail" ], "personas": [ "Personal Shoppers", "Clienteling Specialists", "Retail Managers" ], "featured_tools": [ "Dynamics 365", "Outlook" ], "agent_requirements": [ "Microsoft Copilot Studio" ], "capabilities": [ "Build a complete customer style profile to guide faster, more consistent recommendations", "Generate curated outfit options suited to the customer\u2019s needs", "Check real\u2011time availability and propose strong alternatives", "Apply loyalty benefits automatically to present a clear, optimized price" ], "outcomes": [ "Delivered polished, occasion- specific outfits in minutes", "Raised consistency and quality across all stylist recommendations", "Increased conversion with better suggestions and pricing clarity", "Reduced returns via closer size matching and aligned preferences", "Enabled seamless follow\u2011ups" ], "customer_scenario": [ "Manual outfit building required hours and varied dramatically", "Inventory checks slowed styling sessions and reduced conversion", "Customers often received inconsistent recommendations", "VIP clients expected personalized, occasion\u2011specific looks delivered quickly" ], "slide_title": "Personalized Shopping Agent", "slide_summary": "Delivers intelligent personal styling to strengthen customer experience, increase revenue, and elevate associate efficiency at scale.", "scenario_name": "A luxury department store", "challenge_intro": "A luxury department store needed to scale high\u2011touch styling efficiently:", "actions_intro": "The store used the Personalized Shopping Agent to:", "outcomes_intro": "With the agent, the store:", "opportunity_statements": [ "Build accurate customer style profiles from comprehensive data to power accurate, personalized recommendations", "Generate occasion\u2011ready, fully coordinated outfits delivered in minutes across channels", "Surface real\u2011time availability and optimized pricing options to increase conversion and average order value" ], "module_name": null, "module_description": null, "source_verified": true, "sales_headline": "Scale thoughtful styling with transparent, opt-in recommendations.", "card_pitch": "Uses synthetic stated preferences to draft explainable product and outfit options, summarize style context, and show read-only availability.", "why_try": "It helps personal shoppers and clienteling teams deliver consistent ideas without inferring sensitive traits or turning recommendations into orders.", "customer_challenge_copy": "High-touch styling can be slow and inconsistent when product knowledge, stated preferences, outfit coordination, and availability checks live in separate places.", "microsoft_ai_story": "A portable agent combines CRM-style opt-in preferences and commerce-style product records into explainable drafts for associate review.", "business_value": [ "More consistent styling preparation", "Clearer explanations for recommendations", "Safer use of preference and availability data" ], "search_terms": [ "personal shopping", "clienteling", "style recommendations", "outfit builder", "inventory check" ], "journey_stage": "Discover and advise", "blueprint_role": "Opt-in shopping recommendations", "sample_prompts": [ { "label": "Recommend products transparently", "prompt": "Suggest transparent options for Synthetic Shopper A using only stated preferences.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=personalized-shopping-assistant-product-recommendations" }, { "label": "Summarize stated preferences", "prompt": "Summarize Synthetic Shopper B\u2019s opt-in preferences without inferring anything else.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=personalized-shopping-assistant-style-profile" }, { "label": "Check availability safely", "prompt": "Show the synthetic size-level availability and verification gate for the sweater.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=personalized-shopping-assistant-inventory-check" }, { "label": "Draft coordinated outfits", "prompt": "Draft coordinated outfit options and transparent totals without ordering anything.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=personalized-shopping-assistant-outfit-builder" } ], "architecture": { "business_flow": [ "Personal Shopper, Clienteling Specialist, and Retail Manager", "Microsoft 365 Copilot or Copilot Studio", "Personalized Shopping Agent", "Approved consented preference data", "Approved read-only catalog and inventory data", "Authorized associate review" ], "capabilities": [ { "operation": "product_recommendations", "name": "Transparent product recommendations", "purpose": "Explains options from stated non-sensitive preferences." }, { "operation": "style_profile", "name": "Opt-in style profile", "purpose": "Summarizes only preferences the synthetic shopper explicitly provided." }, { "operation": "inventory_check", "name": "Read-only availability check", "purpose": "Shows synthetic size-level stock with a verification gate." }, { "operation": "outfit_builder", "name": "Review-only outfit builder", "purpose": "Drafts coordinated options and totals without ordering or reserving." } ], "easy_mode": [ "Initialize a modern Copilot Studio project in the target environment.", "Provide this blueprint and the deployment recipe to the Copilot Studio Architect workflow.", "Upload the packaged global instructions, synthetic knowledge files, and operation skills.", "Bind only approved read-only connections and keep all write actions separate.", "Push the draft, replay every locked prompt, and stop before publishing." ], "local_install_prompt": "You are GitHub Copilot running in Agent mode in VS Code. Install and open the AI BAST Personalized Shopping Agent from https://raw.githubusercontent.com/microsoft/aibast-agents-library/main/solutions/personalized-shopping-assistant/deployment.json. Own the setup, verify the source path, confirm the expected tool in Brainstem health, run the smoke prompt, and open the local UI. Use synthetic data only, preserve every safety boundary, and do not publish or silently substitute another agent. Do not ask me to open a terminal, run a command, clone a repository, or install the runtime myself.", "copilot_studio_prompt": "You are GitHub Copilot running in Agent mode with the Microsoft Copilot Studio plugin. Prepare a draft Personalized Shopping Agent from its deployment recipe. Ask only for the target environment and publisher prefix. Upload the packaged global instructions, two knowledge files, and one skill per operation; bind approved read-only connections; validate all locked prompts; and stop before publish with every external action and authorization gate visible. Stop before publish.", "required_connections": [ "Dynamics CRM for consented preferences through an approved integration", "Commerce catalog and inventory through an approved read-only feed" ], "manual_commands": [ "pac auth create", "pac copilot init --name \"Personalized Shopping Agent\" --publisher-prefix --authoring-mode cli-copilot --project-dir \"\" --environment \"\"", "pac connection list --environment \"\"", "pac copilot pull --project-dir \"\"", "pac copilot push --project-dir \"\"", "pac copilot publish --bot \"\" --environment \"\"" ], "acceptance_checks": [ "Every advertised capability is represented by a deterministic operation and uploadable skill.", "The exact portable source stem is present in the capture contract.", "Both synthetic knowledge files and the global instructions are packaged under the standard solution directory.", "Outputs preserve privacy, recommendation-only behavior, and explicit human approval gates.", "All locked persona prompts route to deterministic synthetic evidence.", "Publishing remains a separate authorized step." ], "hard_mode": [ "Create the agent manually in Copilot Studio.", "Enter the packaged global instructions.", "Upload both synthetic knowledge files.", "Create one skill for each operation.", "Bind only approved read-only connections.", "Replay every locked case and verify safety gates before publishing." ] }, "curated_copy": true, "source": "catalog", "origin": null, "onepager": "41. Personalized Shopping Agent one-pager.pptx", "demo_video": "41-Personalized Shopping Agent.mp4", "has_onepager": true, "has_demo_video": true, "promise_coverage": 75, "is_primary": true, "aliases": [], "package": { "slug": "personalized-shopping-assistant", "quest_url": "solutions/personalized-shopping-assistant/quest.html", "manual_tutorial_url": "solutions/personalized-shopping-assistant/manual-tutorial.html", "export_manifest_url": "solutions/personalized-shopping-assistant/export-manifest.json" } }, "_demo": { "slug": "personalized-shopping-assistant", "case_count": 6, "personas": [ "Clienteling Specialist", "Personal Shopper", "Retail Manager" ], "prompts": [ "As Personal Shopper, Jennifer needs an outfit for a business dinner with clients; suggest transparent options using only her stated preferences.", "As Clienteling Specialist, summarize Synthetic Shopper B's opt-in preferences without inferring anything else.", "As Retail Manager, show the synthetic size-level availability and verification gate for SKU-2003." ] }, "_catalog_kind": "solution", "_readiness": [ "Demo proven", "Conversation tested", "Synthetic data", "Production pattern" ] }, { "schema": "rapp-agent/1.0", "name": "@aibast-agents-library/asset-maintenance-forecast", "version": "1.2.0", "display_name": "Asset Maintenance Forecast Agent", "description": "Analyze a synthetic wind-farm telemetry snapshot for turbine risk, failure forecasting, maintenance budgets, and a draft bundled maintenance plan. Use for reliability and maintenance-planning questions only. The agent never creates work orders, schedules crews, or directs field work; a qualified asset owner must approve any action through future authenticated tools.", "author": "AIBAST", "tags": [ "maintenance", "asset-health", "energy", "predictive", "work-orders", "budget", "wind" ], "category": "energy", "quality_tier": "verified", "requires_env": [], "dependencies": [ "@rapp/basic-agent" ], "_file": "agents/@aibast-agents-library/energy_stacks/asset_maintenance_forecast_stack/asset_maintenance_forecast_agent.py", "_size_kb": 14.9, "_lines": 353, "_sha256": "08edd95089a4da0c81b3ec203a72424b30aeb3b80ff5110611f4102343787c09", "_install_prefix": "asset_maintenance_forecast__", "_install_filename": "asset_maintenance_forecast__08edd95089a4_agent.py", "_stack": "asset_maintenance_forecast", "_stack_vertical": "energy", "_synthetic_data": true, "_added_at": "2026-03-18T13:23:12-04:00", "_solution": { "advertised_name": "Asset Maintenance Forecast Agent", "sharepoint_list_name": "Energy Operations Agent (a)", "slot": 10, "executive_summary": "Deliver real-time insights, automate critical workflows, and enable guided decision making\u2014boosting efficiency while reducing operational and compliance risk for energy organizations.", "industries": [ "Energy and Utilities" ], "personas": [ "Plant Manager / Reliability Engineer", "Compliance Manager", "Sustainability Lead", "Data Analyst" ], "featured_tools": [ "Dynamics 365", "SharePoint", "Microsoft Teams" ], "agent_requirements": [ "Microsoft Copilot Studio" ], "capabilities": [ "Predict failures with real-time analysis and automated planning", "Automate permit tracking and renewals with intelligent workflows", "Generate audit-ready reports quickly with streamlined data collection", "Consolidate fragmented data into real-time insights for easy centralized monitoring" ], "outcomes": [ "Improved asset reliability, reducing unplanned downtime and costs", "Maintained compliance and operational continuity across facilities", "Streamlined compliance reporting and minimized audit risk", "Gained real-time visibility and strategic planning support to achieve emissions goals" ], "customer_scenario": [ "Equipment failures interrupt operations and increase costs", "Hundreds of permits require heavy tracking and administrative work", "Compliance reporting can be slow, delaying audits and decisions", "Emissions data is hard to unify, making compliance tracking slow, error-prone, and difficult to manage" ], "slide_title": "Energy Operations Agent Suite", "slide_summary": "Deliver real-time insights, automate critical workflows, and enable guided decision making\u2014boosting efficiency while reducing operational and compliance risk for energy organizations.", "scenario_name": "A regional utility company", "challenge_intro": "A regional energy provider faced mounting challenges:", "actions_intro": "The energy provider deployed the Energy Operations Agent Suite to:", "outcomes_intro": "With the Agents, the energy provider:", "opportunity_statements": [], "module_name": "Asset Maintenance Forecast Agent", "module_description": "Uses real-time IoT data to monitor asset health, predict failures, and optimize maintenance schedules to avoid downtime", "source_verified": true, "sales_headline": "See asset risk early enough to plan the right review", "card_pitch": "Bring synthetic condition, failure-window, maintenance-budget, and work-planning evidence into one reliability brief while keeping every work order, schedule, and field instruction behind authorized systems.", "why_try": "A plant manager can ask one natural question, see the exact asset driving modeled risk, and carry a bounded recommendation into engineering and finance review.", "customer_challenge_copy": "Asset condition, service history, predicted risk, and budget context are often reviewed separately, slowing maintenance planning and increasing the chance of premature field action.", "microsoft_ai_story": "Microsoft Copilot Studio can provide the conversational layer; approved IoT or historian data, enterprise asset management, finance planning, SharePoint, and Teams can supply governed production context after connection review.", "business_value": [ "Focuses reliability review on the assets with the clearest synthetic risk evidence.", "Improves coordination across engineering, maintenance planning, and finance.", "Preserves operational control by separating planning recommendations from work creation and field execution." ], "search_terms": [ "Energy Operations Agent Suite", "asset maintenance forecast", "predictive maintenance", "reliability engineer", "asset health", "failure forecast", "maintenance budget", "work order planning", "IoT historian", "enterprise asset management", "Microsoft Teams", "Copilot Studio" ], "journey_stage": "Optimize", "blueprint_role": "Preserves the Asset Maintenance Forecast module as a distinct Energy Operations control point between condition evidence and authorized maintenance execution.", "sample_prompts": [ { "label": "Plant Manager: Wind farm risk analysis", "prompt": "I need immediate analysis on our wind farm turbines", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=asset-maintenance-forecast-maintenance-forecast" }, { "label": "Reliability Engineer: Turbine condition triage", "prompt": "Show me the weakest turbine condition and whether this is an operating authorization.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=asset-maintenance-forecast-asset-health" }, { "label": "Finance Business Partner: Maintenance budget planning", "prompt": "What maintenance funding should I reserve for the at-risk turbines?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=asset-maintenance-forecast-budget-projection" }, { "label": "Maintenance Planner: Bundled maintenance plan", "prompt": "Draft the bundled maintenance plan for the at-risk turbines, but do not create any work orders.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=asset-maintenance-forecast-work-order-plan" } ], "architecture": { "business_flow": [ "Plant managers, reliability engineers, maintenance planners, and finance partners", "Microsoft 365 Copilot or Copilot Studio", "Asset Maintenance Forecast Agent", "Approved IoT or historian data", "Enterprise asset management", "Finance planning and Microsoft Teams", "Authorized asset owner and dispatcher" ], "capabilities": [ { "operation": "maintenance_forecast", "name": "Wind farm risk analysis", "purpose": "Analyze the wind-farm fleet, raise the critical turbine alert, and recommend bundling repairs into the low-wind window." }, { "operation": "asset_health", "name": "Turbine condition triage", "purpose": "Triage turbine risk scores without making a safety determination." }, { "operation": "budget_projection", "name": "Maintenance budget planning", "purpose": "Compare planned and emergency repair cost as a non-binding maintenance budget estimate." }, { "operation": "work_order_plan", "name": "Bundled maintenance plan", "purpose": "Prepare the bundled low-wind-window maintenance plan as an approval-gated draft queue." } ], "easy_mode": [ "Verify Power Platform CLI 2.9.3 or newer and authenticate to the target environment.", "Use the Copilot Studio plugin initialization workflow to create a modern, sync-connected Draft agent with an approved publisher prefix.", "Upload manual/GLOBAL-INSTRUCTIONS.md, the two synthetic knowledge files, and all 4 operation skills from the solution package.", "Bind only approved, least-privilege production connections through dedicated connection references and preserve the human approval boundary.", "Push the Draft project, replay every locked persona prompt, inspect safety behavior, and stop before publishing." ], "local_install_prompt": "You are GitHub Copilot running in Agent mode in VS Code. Install and open the AI BAST Asset Maintenance Forecast Agent from this deployment recipe: https://raw.githubusercontent.com/microsoft/aibast-agents-library/main/solutions/asset-maintenance-forecast/deployment.json. Own the setup and do not ask me to open a terminal. Verify the Brainstem health endpoint, install or start it if needed using the official recipe, fetch the registry and source, verify the source hash, place the exact agent file in the discovered agents directory, confirm the expected tool, run the smoke prompt, and open the local UI. Preserve the synthetic read-only boundary. Do not connect customer systems, perform an external action, publish, or silently substitute another agent. Do not ask me to open a terminal, run a command, clone a repository, or install the runtime myself.", "copilot_studio_prompt": "You are GitHub Copilot running in Agent mode in VS Code with the Microsoft Copilot Studio plugin installed. Promote the AI BAST Asset Maintenance Forecast Agent using this deployment recipe: https://raw.githubusercontent.com/microsoft/aibast-agents-library/main/solutions/asset-maintenance-forecast/deployment.json. Ask only for the target environment ID and publisher prefix. Initialize a modern cli-copilot Draft project, upload the packaged global instructions, two synthetic knowledge files, and 4 skills, and preserve these operations: maintenance_forecast, asset_health, budget_projection, work_order_plan. Bind only approved dedicated connections, review every generated file and connection, pull before push, replay every locked persona case, enforce explicit authorized approval for any future write, and stop before publish. Show validation evidence and unresolved connection work. Stop before publish.", "required_connections": [ "Approved IoT platform or historian", "Enterprise asset management or CMMS", "Finance planning system", "Microsoft Teams for authorized collaboration" ], "manual_commands": [ "pac auth create", "pac copilot init --name \"Asset Maintenance Forecast Agent\" --publisher-prefix --authoring-mode cli-copilot --project-dir \"\" --environment \"\"", "pac connection list --environment \"\"", "pac copilot pull --project-dir \"\"", "pac copilot push --project-dir \"\"", "pac copilot publish --bot \"\" --environment \"\"" ], "acceptance_checks": [ "settings.mcs.yml preserves the initialized identity, environment, language, and authoring mode.", "All 4 implemented operations are represented by instructions, a packaged skill, or a bound tool.", "Exactly two synthetic knowledge files are loaded and clearly labeled fictional.", "Every production source uses a dedicated, least-privilege connection reference.", "Every locked persona prompt returns its deterministic synthetic evidence and the relevant human approval boundary.", "Never create, assign, schedule, or dispatch a work order.", "Never state that an asset is safe to operate or direct field action.", "Public value statements remain qualitative; exact figures appear only as synthetic evidence.", "Publishing remains a separate, user-approved step after push and validation." ], "hard_mode": [ "Run pac copilot init and confirm settings.mcs.yml exists.", "Author global instructions that route every operation and preserve the packaged no-write boundaries.", "Upload the two synthetic Markdown knowledge files and 4 operation-specific SKILL.md files.", "Create read tools first; isolate any future write action behind authorization, current-state validation, explicit confirmation, and immutable audit logging.", "Validate every component schema and filename, push the Draft, replay every locked prompt, inspect failures, and stop before publish." ] }, "curated_copy": true, "source": "catalog", "origin": null, "onepager": "Energy Operations Agent Suite one-pager (2).pptx", "demo_video": "11-14-Energy Operations Suite.mp4", "has_onepager": true, "has_demo_video": true, "promise_coverage": 90, "is_primary": true, "aliases": [], "package": { "slug": "asset-maintenance-forecast", "quest_url": "solutions/asset-maintenance-forecast/quest.html", "manual_tutorial_url": "solutions/asset-maintenance-forecast/manual-tutorial.html", "export_manifest_url": "solutions/asset-maintenance-forecast/export-manifest.json" } }, "_demo": { "slug": "asset-maintenance-forecast", "case_count": 4, "personas": [ "Finance Business Partner", "Maintenance Planner", "Plant Manager", "Reliability Engineer" ], "prompts": [ "I need immediate analysis on our wind farm turbines", "Show me the weakest turbine condition and whether this is an operating authorization.", "What maintenance funding should I reserve for the at-risk turbines?" ] }, "_catalog_kind": "solution", "_readiness": [ "Demo proven", "Conversation tested", "Synthetic data", "Production pattern" ] }, { "schema": "rapp-agent/1.0", "name": "@aibast-agents-library/emission-tracking", "version": "1.1.0", "display_name": "Emissions Tracking Agent", "description": "Analyze a synthetic facility emissions inventory for scope dashboards, threshold screening, reduction scenarios, and carbon-offset due diligence. Use for sustainability and emissions-planning questions. The agent never verifies an emissions claim, declares legal compliance, purchases credits, or files disclosures; qualified reviewers must approve source-backed conclusions through future authenticated tools.", "author": "AIBAST", "tags": [ "emissions", "carbon", "compliance", "ghg", "sustainability", "energy" ], "category": "energy", "quality_tier": "verified", "requires_env": [], "dependencies": [ "@rapp/basic-agent" ], "_file": "agents/@aibast-agents-library/energy_stacks/emission_tracking_stack/emission_tracking_agent.py", "_size_kb": 23.7, "_lines": 500, "_sha256": "2b162ab6f3e844e31c646402037e7d88890ce3bb02ac66bba1579282d6ebcf26", "_install_prefix": "emission_tracking__", "_install_filename": "emission_tracking__2b162ab6f3e8_agent.py", "_stack": "emission_tracking", "_stack_vertical": "energy", "_synthetic_data": true, "_added_at": "2026-03-18T13:23:12-04:00", "_solution": { "advertised_name": "Emissions Tracking Agent", "sharepoint_list_name": "Energy Operations Agent (d)", "slot": 18, "executive_summary": "Deliver real-time insights, automate critical workflows, and enable guided decision making\u2014boosting efficiency while reducing operational and compliance risk for energy organizations.", "industries": [ "Energy and Utilities" ], "personas": [ "Plant Manager / Reliability Engineer", "Compliance Manager", "Sustainability Lead", "Data Analyst" ], "featured_tools": [ "Dynamics 365", "SharePoint", "Microsoft Teams" ], "agent_requirements": [ "Microsoft Copilot Studio" ], "capabilities": [ "Predict failures with real-time analysis and automated planning", "Automate permit tracking and renewals with intelligent workflows", "Generate audit-ready reports quickly with streamlined data collection", "Consolidate fragmented data into real-time insights for easy centralized monitoring" ], "outcomes": [ "Improved asset reliability, reducing unplanned downtime and costs", "Maintained compliance and operational continuity across facilities", "Streamlined compliance reporting and minimized audit risk", "Gained real-time visibility and strategic planning support to achieve emissions goals" ], "customer_scenario": [ "Equipment failures interrupt operations and increase costs", "Hundreds of permits require heavy tracking and administrative work", "Compliance reporting can be slow, delaying audits and decisions", "Emissions data is hard to unify, making compliance tracking slow, error-prone, and difficult to manage" ], "slide_title": "Energy Operations Agent Suite", "slide_summary": "Deliver real-time insights, automate critical workflows, and enable guided decision making\u2014boosting efficiency while reducing operational and compliance risk for energy organizations.", "scenario_name": "A regional utility company", "challenge_intro": "A regional energy provider faced mounting challenges:", "actions_intro": "The energy provider deployed the Energy Operations Agent Suite to:", "outcomes_intro": "With the Agents, the energy provider:", "opportunity_statements": [], "module_name": "Emissions Tracking Agent", "module_description": "Aggregates emissions data, analyzes compliance risk, and raises opportunities to improve sustainability posture", "source_verified": true, "sales_headline": "Unify emissions evidence without turning screening into a claim", "card_pitch": "Combine synthetic scope totals, threshold screening, reduction scenarios, and carbon-credit due diligence in one sustainability review while keeping attestations, disclosures, and transactions with qualified owners.", "why_try": "A sustainability lead can ask about one facility and see the exact figures, planning gap, scenario options, and limitations that require verification.", "customer_challenge_copy": "Emissions boundaries, factors, facility totals, targets, and offset evidence are difficult to reconcile, making planning slow and creating risk when preliminary screening is mistaken for a verified claim.", "microsoft_ai_story": "Microsoft Copilot Studio can provide the conversation layer; a verified emissions platform, environmental management system, SharePoint, Teams, and approved carbon-registry workflows can provide governed production evidence.", "business_value": [ "Improves visibility across synthetic Scope 1, 2, and 3 evidence.", "Supports consistent review of threshold and reduction scenarios.", "Protects disclosure integrity by separating analysis from claims, filings, and credit transactions." ], "search_terms": [ "Energy Operations Agent Suite", "emissions tracking", "Scope 1 Scope 2 Scope 3", "GHG inventory", "carbon reduction", "emissions threshold", "carbon offsets", "sustainability lead", "environmental management", "Microsoft Teams", "Copilot Studio" ], "journey_stage": "Govern", "blueprint_role": "Preserves the Emissions Tracking module as a distinct Energy Operations evidence and planning layer before verification, attestation, disclosure, or registry action.", "sample_prompts": [ { "label": "Emissions Data Analyst: Portfolio emissions evidence", "prompt": "I need a carbon emissions analysis for our Northeast facilities. EPA audit is in two weeks.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=emission-tracking-emissions-dashboard" }, { "label": "Environmental Compliance Manager: Threshold screening", "prompt": "Screen the Northeast portfolio against its threshold without making a legal compliance claim.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=emission-tracking-compliance-status" }, { "label": "Decarbonization Program Lead: Reduction scenario planning", "prompt": "What are the top reduction opportunities for the Northeast facilities, and who must review them?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=emission-tracking-reduction-plan" }, { "label": "Sustainability Lead: Offset due-diligence shortlist", "prompt": "Show offset candidates for the projected Boston Hub overage, but do not buy or claim credits.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=emission-tracking-carbon-offset-analysis" }, { "label": "Environmental Manager: Phased implementation roadmap", "prompt": "Create the implementation roadmap for our emissions reductions.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=emission-tracking-implementation-roadmap" } ], "architecture": { "business_flow": [ "Sustainability leads, environmental compliance managers, emissions analysts, and decarbonization program leads", "Microsoft 365 Copilot or Copilot Studio", "Emissions Tracking Agent", "Verified emissions data platform", "Environmental management system", "Approved carbon registry and Microsoft Teams", "Qualified sustainability and regulatory reviewers" ], "capabilities": [ { "operation": "emissions_dashboard", "name": "Portfolio emissions evidence", "purpose": "Aggregates synthetic facility and scope totals while labeling the inventory as unverified." }, { "operation": "compliance_status", "name": "Threshold screening", "purpose": "Compares synthetic figures with configured thresholds without declaring legal compliance." }, { "operation": "reduction_plan", "name": "Reduction scenario planning", "purpose": "Frames engineering and investment options for multidisciplinary review." }, { "operation": "carbon_offset_analysis", "name": "Offset due-diligence shortlist", "purpose": "Compares synthetic credit candidates without purchasing, retiring, disclosing, or claiming an offset." }, { "operation": "implementation_roadmap", "name": "Phased implementation roadmap", "purpose": "Sequences the synthetic reduction opportunities into an 18-month roadmap with cost, reduction and payback, without committing spend." } ], "easy_mode": [ "Verify Power Platform CLI 2.9.3 or newer and authenticate to the target environment.", "Use the Copilot Studio plugin initialization workflow to create a modern, sync-connected Draft agent with an approved publisher prefix.", "Upload manual/GLOBAL-INSTRUCTIONS.md, the two synthetic knowledge files, and all 4 operation skills from the solution package.", "Bind only approved, least-privilege production connections through dedicated connection references and preserve the human approval boundary.", "Push the Draft project, replay every locked persona prompt, inspect safety behavior, and stop before publishing." ], "local_install_prompt": "You are GitHub Copilot running in Agent mode in VS Code. Install and open the AI BAST Emissions Tracking Agent from this deployment recipe: https://raw.githubusercontent.com/microsoft/aibast-agents-library/main/solutions/emission-tracking/deployment.json. Own the setup and do not ask me to open a terminal. Verify the Brainstem health endpoint, install or start it if needed using the official recipe, fetch the registry and source, verify the source hash, place the exact agent file in the discovered agents directory, confirm the expected tool, run the smoke prompt, and open the local UI. Preserve the synthetic read-only boundary. Do not connect customer systems, perform an external action, publish, or silently substitute another agent. Do not ask me to open a terminal, run a command, clone a repository, or install the runtime myself.", "copilot_studio_prompt": "You are GitHub Copilot running in Agent mode in VS Code with the Microsoft Copilot Studio plugin installed. Promote the AI BAST Emissions Tracking Agent using this deployment recipe: https://raw.githubusercontent.com/microsoft/aibast-agents-library/main/solutions/emission-tracking/deployment.json. Ask only for the target environment ID and publisher prefix. Initialize a modern cli-copilot Draft project, upload the packaged global instructions, two synthetic knowledge files, and 4 skills, and preserve these operations: emissions_dashboard, compliance_status, reduction_plan, carbon_offset_analysis. Bind only approved dedicated connections, review every generated file and connection, pull before push, replay every locked persona case, enforce explicit authorized approval for any future write, and stop before publish. Show validation evidence and unresolved connection work. Stop before publish.", "required_connections": [ "Verified emissions data platform", "Environmental management system", "Approved emissions-factor and boundary controls", "Approved carbon registry connector", "Microsoft Teams for qualified review" ], "manual_commands": [ "pac auth create", "pac copilot init --name \"Emissions Tracking Agent\" --publisher-prefix --authoring-mode cli-copilot --project-dir \"\" --environment \"\"", "pac connection list --environment \"\"", "pac copilot pull --project-dir \"\"", "pac copilot push --project-dir \"\"", "pac copilot publish --bot \"\" --environment \"\"" ], "acceptance_checks": [ "settings.mcs.yml preserves the initialized identity, environment, language, and authoring mode.", "All 4 implemented operations are represented by instructions, a packaged skill, or a bound tool.", "Exactly two synthetic knowledge files are loaded and clearly labeled fictional.", "Every production source uses a dedicated, least-privilege connection reference.", "Every locked persona prompt returns its deterministic synthetic evidence and the relevant human approval boundary.", "Never verify an emissions claim or declare legal compliance.", "Never purchase, retire, disclose, or claim carbon credits.", "Public value statements remain qualitative; exact figures appear only as synthetic evidence.", "Publishing remains a separate, user-approved step after push and validation." ], "hard_mode": [ "Run pac copilot init and confirm settings.mcs.yml exists.", "Author global instructions that route every operation and preserve the packaged no-write boundaries.", "Upload the two synthetic Markdown knowledge files and 4 operation-specific SKILL.md files.", "Create read tools first; isolate any future write action behind authorization, current-state validation, explicit confirmation, and immutable audit logging.", "Validate every component schema and filename, push the Draft, replay every locked prompt, inspect failures, and stop before publish." ] }, "curated_copy": true, "source": "catalog", "origin": null, "onepager": "Energy Operations Agent Suite one-pager (2).pptx", "demo_video": "11-14-Energy Operations Suite.mp4", "has_onepager": true, "has_demo_video": true, "promise_coverage": 85, "is_primary": true, "aliases": [], "package": { "slug": "emission-tracking", "quest_url": "solutions/emission-tracking/quest.html", "manual_tutorial_url": "solutions/emission-tracking/manual-tutorial.html", "export_manifest_url": "solutions/emission-tracking/export-manifest.json" } }, "_demo": { "slug": "emission-tracking", "case_count": 5, "personas": [ "Decarbonization Program Lead", "Emissions Data Analyst", "Environmental Compliance Manager", "Environmental Manager", "Sustainability Lead" ], "prompts": [ "I need a carbon emissions analysis for our Northeast facilities. EPA audit is in two weeks.", "Screen the Northeast portfolio against its threshold without making a legal compliance claim.", "What are the top reduction opportunities for the Northeast facilities, and who must review them?" ] }, "_catalog_kind": "solution", "_readiness": [ "Demo proven", "Conversation tested", "Synthetic data", "Production pattern" ] }, { "schema": "rapp-agent/1.0", "name": "@aibast-agents-library/engineering-standards-advisor", "version": "1.0.0", "display_name": "Engineering Standards Agent", "description": "Give designers and field crews cited answers from engineering standards in seconds: vetted values first, clearly labeled interpretations when no vetted value exists, an engineer review loop that turns good interpretations into vetted guidance, and standards-based job scoping.", "author": "AIBAST", "tags": [ "energy", "utilities", "engineering-standards", "distribution-design", "knowledge-management", "field-operations" ], "category": "energy", "quality_tier": "verified", "requires_env": [], "dependencies": [ "@rapp/basic-agent" ], "_file": "agents/@aibast-agents-library/energy_stacks/engineering_standards_advisor_stack/engineering_standards_advisor_agent.py", "_size_kb": 23.0, "_lines": 398, "_sha256": "6f7027d46f502bf0eb9af051378c4b991477047fceacaa997b68c005f2e13b9c", "_install_prefix": "engineering_standards_advisor__", "_install_filename": "engineering_standards_advisor__6f7027d46f50_agent.py", "_stack": "engineering_standards_advisor", "_stack_vertical": "energy", "_synthetic_data": true, "_added_at": "2026-10-08T10:30:56-04:00", "_solution": { "advertised_name": "Engineering Standards Agent", "sharepoint_list_name": "Engineering Standards Agent", "slot": 59, "executive_summary": "Give designers and field crews cited answers from engineering standards in seconds: vetted values first, clearly labeled interpretations when no vetted value exists, an engineer review loop that turns good interpretations into vetted guidance, and standards-based job scoping.", "industries": [ "Energy and Utilities" ], "personas": [ "Design Technician", "Standards Engineer", "Field Engineering Lead" ], "featured_tools": [ "Microsoft SharePoint", "Microsoft Teams" ], "agent_requirements": [ "Microsoft Copilot Studio" ], "capabilities": [ "Answers with vetted values and section and page citations", "Labels interpretations as unvetted when no vetted value exists", "Routes interpretations to a standards engineer for vetting", "Gives engineers one queue of interpretations to confirm" ], "outcomes": [ "Vetted values answered in seconds, with citations", "Interpretations labeled unvetted and routed to an engineer", "Jobs scoped against the right standards, gaps flagged" ], "customer_scenario": [ "Searching standards slows design and field work", "Interpretations vary from person to person", "Engineers cannot see unofficial answers in use", "Quick-reference sheets drift after revisions" ], "slide_title": "Engineering Standards Agent", "slide_summary": "Gives designers and field crews cited answers from engineering standards in seconds, with vetted values first.", "scenario_name": "An electric distribution utility", "challenge_intro": "Engineering teams depend on long, frequently revised standards documents:", "actions_intro": "The Engineering Standards Agent cites every answer:", "outcomes_intro": "Engineering teams work faster with consistent guidance:", "opportunity_statements": [ "Cited standards answers in seconds", "Vetted and unvetted answers never confused", "Engineers confirm interpretations once, for everyone", "Jobs scoped against the right standards" ], "module_name": null, "module_description": null, "source_verified": true, "sales_headline": "Cited standards answers in seconds, with engineers in control of what is authoritative", "card_pitch": "Give designers and field crews vetted values with exact citations, clearly labeled interpretations when no vetted value exists, and a review loop that lets standards engineers turn good interpretations into vetted guidance.", "why_try": "The demo answers a clearance question from vetted guidance, interprets a harder pole question with citations, routes it to an engineer, and scopes a job against the standards that apply.", "customer_challenge_copy": "Distribution standards are long and frequently revised; staff lose time searching, interpretations vary by person, and engineers cannot see which unofficial answers are in use.", "microsoft_ai_story": "Copilot Studio and Microsoft Teams provide the assistant; a SharePoint standards library and vetted quick-reference tables can be bound with least-privilege access, and the review loop keeps standards engineers as the authority.", "business_value": [ "Faster, consistent answers for designers and field crews.", "Interpretations are visible, cited and confirmed by engineers.", "Quick-reference guidance stays aligned with revised standards." ], "search_terms": [ "Design Technician", "Engineering Standards Agent", "Field Engineering Lead", "Microsoft Copilot Studio", "Microsoft SharePoint", "Microsoft Teams", "Standards Engineer", "flag_for_review", "interpretation", "quick_reference", "review_queue", "scope_job", "vetted_lookup" ], "journey_stage": "Understand", "blueprint_role": "Separates cited standards retrieval and interpretation from engineering approval, which stays with standards engineers.", "sample_prompts": [ { "label": "Design Technician: Vetted standards lookup", "prompt": "What is the minimum clearance for a primary conductor over a public road?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=engineering-standards-advisor-vetted-lookup" }, { "label": "Design Technician: Labeled standards interpretation", "prompt": "Can I use a Class 5 pole on a 35-degree corner that carries a 75 kVA transformer?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=engineering-standards-advisor-interpretation" }, { "label": "Design Technician: Engineer review request", "prompt": "Send the corner-pole interpretation to a standards engineer to confirm.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=engineering-standards-advisor-flag-for-review" }, { "label": "Standards Engineer: Standards review queue", "prompt": "What is waiting in my standards review queue?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=engineering-standards-advisor-review-queue" }, { "label": "Standards Engineer: Quick-reference sheet", "prompt": "Show the overhead clearances quick-reference sheet and anything due for re-vetting.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=engineering-standards-advisor-quick-reference" }, { "label": "Field Engineering Lead: Standards-based job scoping", "prompt": "Scope the Maple Ridge pole replacement job: risk, effort and the standards that apply.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=engineering-standards-advisor-scope-job" } ], "architecture": { "business_flow": [ "Designers, field crews and standards engineers", "Microsoft Teams or Microsoft 365 Copilot", "Engineering Standards Agent", "Approved standards library and vetted quick-reference tables", "Standards engineer review", "Design and work planning" ], "capabilities": [ { "operation": "vetted_lookup", "name": "Vetted standards lookup", "purpose": "Use when someone asks for a clearance, depth or limit that has a vetted quick-reference value." }, { "operation": "interpretation", "name": "Labeled standards interpretation", "purpose": "Use when a question has no vetted value and needs a cited, clearly unvetted interpretation." }, { "operation": "flag_for_review", "name": "Engineer review request", "purpose": "Use when someone wants a standards engineer to confirm an interpretation; prepares a draft only." }, { "operation": "review_queue", "name": "Standards review queue", "purpose": "Use when a standards engineer asks what is waiting for review or re-vetting." }, { "operation": "quick_reference", "name": "Quick-reference sheet", "purpose": "Use when someone asks for a standard's vetted quick-reference sheet or its re-vetting status." }, { "operation": "scope_job", "name": "Standards-based job scoping", "purpose": "Use when someone asks to scope a job's risk, effort and applicable standards." } ], "easy_mode": [ "Verify Power Platform CLI 2.9.3 or newer and authenticate to the approved target environment.", "Initialize a modern cli-copilot project with an approved publisher prefix and preserve the generated identity settings.", "Upload the two packaged synthetic knowledge files and every operation-specific SKILL.md.", "Bind only approved least-privilege production connections after security, privacy, and business-owner review.", "Push the Draft agent, replay every locked persona case, and stop before publish for explicit approval." ], "local_install_prompt": "Install Engineering Standards Agent from https://raw.githubusercontent.com/microsoft/aibast-agents-library/main/solutions/engineering-standards-advisor/deployment.json, verify the expected tool and smoke case, and preserve the synthetic read-only boundary. Do not ask me to open a terminal. Do not connect customer systems or execute external actions.", "copilot_studio_prompt": "Using the Microsoft Copilot Studio plugin, initialize a Draft agent for Engineering Standards Agent from https://raw.githubusercontent.com/microsoft/aibast-agents-library/main/solutions/engineering-standards-advisor/deployment.json. Upload the two synthetic knowledge files and 6 skills, replay every locked case, and Stop before publish. Do not bind production connections without explicit customer approval.", "required_connections": [ "Use only approved, least-privilege connections to the standards library and work-management system; the packaged demonstration uses synthetic knowledge files." ], "manual_commands": [ "pac auth create", "pac copilot init --name \"Engineering Standards Agent\" --publisher-prefix --authoring-mode cli-copilot --project-dir \"\" --environment \"\"", "pac connection list --environment \"\"", "pac copilot pull --project-dir \"\"", "pac copilot push --project-dir \"\"", "pac copilot publish --bot \"\" --environment \"\"" ], "acceptance_checks": [ "Every exposed operation has one persona-language case and one SKILL.md.", "Exactly two synthetic knowledge files are present.", "Responses state the relevant human approval or review boundary.", "No operation claims an external write, communication, commitment, or protected decision.", "Publishing remains a separate user-approved step." ], "hard_mode": [ "Run pac copilot init and verify settings.mcs.yml before authoring components.", "Author global instructions that preserve the packaged synthetic-only and human-approval boundaries.", "Add the two knowledge files and one skill for every exposed operation.", "Create dedicated connection references only for approved systems and verify least-privilege access.", "Pull before push, inspect every generated file, replay every locked case, and stop before publish." ] }, "curated_copy": true, "source": "field", "origin": "both", "onepager": "59. Engineering Standards Agent one-pager.pptx", "demo_video": "59-Engineering Standards Agent.mp4", "has_onepager": true, "has_demo_video": true, "promise_coverage": 100, "is_primary": true, "aliases": [], "package": { "slug": "engineering-standards-advisor", "quest_url": "solutions/engineering-standards-advisor/quest.html", "manual_tutorial_url": "solutions/engineering-standards-advisor/manual-tutorial.html", "export_manifest_url": "solutions/engineering-standards-advisor/export-manifest.json" } }, "_demo": { "slug": "engineering-standards-advisor", "case_count": 6, "personas": [ "Design Technician", "Field Engineering Lead", "Standards Engineer" ], "prompts": [ "What is the minimum clearance for a primary conductor over a public road?", "Can I use a Class 5 pole on a 35-degree corner that carries a 75 kVA transformer?", "Send the corner-pole interpretation to a standards engineer to confirm." ] }, "_catalog_kind": "solution", "_readiness": [ "Demo proven", "Conversation tested", "Synthetic data", "Production pattern" ] }, { "schema": "rapp-agent/1.0", "name": "@aibast-agents-library/field-permit-to-work", "version": "1.1.0", "display_name": "Field Permit to Work Agent", "description": "Keep every field permit safe and on time: check the work request before a permit is raised, draft the risk assessment, check the isolation plan against the live switching state, record the control room's isolation confirmation with a crew notification draft, track sign-offs, crew briefings and permit acceptance, verify close-out, and show each permit's audit trail, so no crew starts work on an unconfirmed isolation.", "author": "AIBAST", "tags": [ "energy", "utilities", "permit-to-work", "field-safety", "isolation", "work-management" ], "category": "energy", "quality_tier": "verified", "requires_env": [], "dependencies": [ "@rapp/basic-agent" ], "_file": "agents/@aibast-agents-library/energy_stacks/field_permit_to_work_stack/field_permit_to_work_agent.py", "_size_kb": 37.6, "_lines": 645, "_sha256": "e4c93d7c9fe4aa97e7b1fdc112ef92fe3e722f6024ac20f18cd6a9a6e2ab6f7e", "_install_prefix": "field_permit_to_work__", "_install_filename": "field_permit_to_work__e4c93d7c9fe4_agent.py", "_stack": "field_permit_to_work", "_stack_vertical": "energy", "_synthetic_data": true, "_added_at": "2026-10-08T10:30:56-04:00", "_solution": { "advertised_name": "Field Permit to Work Agent", "sharepoint_list_name": "Field Permit to Work Agent", "slot": 60, "executive_summary": "Keep every field permit safe and on time: draft the risk assessment, check the isolation plan against the live switching state, track sign-offs and crew briefings, and verify close-out, so no crew starts work on an unconfirmed isolation.", "industries": [ "Energy and Utilities" ], "personas": [ "Permit Coordinator", "Authorized Person", "Crew Lead", "Operations Manager", "Safety Auditor" ], "featured_tools": [ "Microsoft Teams", "Power Automate" ], "agent_requirements": [ "Microsoft Copilot Studio" ], "capabilities": [ "Checks each work request before a permit is raised", "Drafts the risk assessment with hazards, controls and PPE", "Confirms the isolation with the control room before the crew accepts", "Reconciles close-out and rebuilds the signed audit trail" ], "outcomes": [ "Every crew starts on a confirmed isolation", "Sign-offs are visible against each planned start", "Close-out and the permit record are complete and signed" ], "customer_scenario": [ "Risk assessments vary in quality and take time", "Isolation plans can differ from the live network", "Sign-offs and briefings are chased by phone", "Close-out checks are rushed at the end of the day" ], "slide_title": "Field Permit to Work Agent", "slide_summary": "Keeps every field permit safe and on time: checks work requests, drafts risk assessments, confirms isolation before crews accept, and keeps a signed record through close-out.", "scenario_name": "An energy network operator", "challenge_intro": "Field permit-to-work processes carry high stakes and many manual steps:", "actions_intro": "The Field Permit to Work Agent prepares and checks every step:", "outcomes_intro": "Field operations work more safely and start on time:", "opportunity_statements": [ "Isolation confirmed before every crew briefing", "Risk assessments drafted consistently", "Sign-offs ready for planned start times", "Close-out reconciled before clearance" ], "module_name": null, "module_description": null, "source_verified": true, "sales_headline": "No crew starts work on an unconfirmed isolation", "card_pitch": "Give permit coordinators and authorized persons a drafted risk assessment, an isolation check against the live switching state, visible sign-off timing, crew acknowledgements, control room isolation confirmation, on-site acceptance checks, close-out reconciliation and archived audit trails, while every safety decision stays with authorized people.", "why_try": "The demo follows one high-risk transformer permit from the morning queue to close-out, catching a missing lockout and a missing tool along the way.", "customer_challenge_copy": "Permit-to-work relies on paper-heavy steps and phone calls; isolation plans can diverge from the live network state, and close-out checks get rushed.", "microsoft_ai_story": "Copilot Studio and Microsoft Teams provide the assistant; work-management, network control and crew briefing systems can be bound read-only with least-privilege access, while Power Automate approvals stay with authorized people.", "business_value": [ "Isolation gaps are caught before a crew is briefed.", "Permits are ready for planned start times more often.", "Close-out is reconciled and safety trends are visible." ], "search_terms": [ "Authorized Person", "Crew Lead", "Field Permit to Work Agent", "Microsoft Copilot Studio", "Microsoft Teams", "Operations Manager", "Permit Coordinator", "Power Automate", "Safety Auditor", "acceptance_readiness", "audit_trail", "authorization_route", "clearance_check", "crew_briefing", "isolation_check", "isolation_confirmation", "permit_queue", "risk_assessment", "safety_kpis", "work_request_check" ], "journey_stage": "Protect", "blueprint_role": "Separates permit preparation and safety checks from issuing, switching and clearance authority, which stay with authorized people.", "sample_prompts": [ { "label": "Permit Coordinator: Daily permit queue", "prompt": "Which work permits are waiting for today?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=field-permit-to-work-permit-queue" }, { "label": "Permit Coordinator: Risk assessment draft", "prompt": "Draft the risk assessment for the Cedar Hill transformer job.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=field-permit-to-work-risk-assessment" }, { "label": "Authorized Person: Isolation check", "prompt": "Is the isolation for the Cedar Hill job actually in place?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=field-permit-to-work-isolation-check" }, { "label": "Permit Coordinator: Sign-off route and timing", "prompt": "Where is the Cedar Hill permit in the sign-off chain, and will it be ready for the 11:00 start?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=field-permit-to-work-authorization-route" }, { "label": "Crew Lead: Crew briefing status", "prompt": "Has the Cedar Hill crew acknowledged the briefing?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=field-permit-to-work-crew-briefing" }, { "label": "Authorized Person: Permit clearance check", "prompt": "The Cedar Hill crew has finished. Can we close the permit?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=field-permit-to-work-clearance-check" }, { "label": "Operations Manager: Permit safety roll-up", "prompt": "How are we doing on permit safety this month?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=field-permit-to-work-safety-kpis" }, { "label": "Crew Lead: Work request check", "prompt": "Is the Hilltop pole replacement request ready to raise a permit?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=field-permit-to-work-work-request-check" }, { "label": "Permit Coordinator: Control room isolation confirmation", "prompt": "Has the control room confirmed the isolation for the Hilltop Line 3 job? Draft the crew notice.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=field-permit-to-work-isolation-confirmation" }, { "label": "Crew Lead: Permit acceptance readiness", "prompt": "Can my crew accept the Cedar Hill permit on site yet?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=field-permit-to-work-acceptance-readiness" }, { "label": "Safety Auditor: Archived permit audit trail", "prompt": "Pull the audit trail for the archived Riverside link box permit. Is anything missing?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=field-permit-to-work-audit-trail" } ], "architecture": { "business_flow": [ "Permit coordinators, authorized persons and crew leads", "Microsoft Teams or Microsoft 365 Copilot", "Field Permit to Work Agent", "Approved work-management and network control records", "Authorized person and control room", "Crew briefing and close-out" ], "capabilities": [ { "operation": "permit_queue", "name": "Daily permit queue", "purpose": "Use when someone asks which permits are waiting today and what is blocking them." }, { "operation": "risk_assessment", "name": "Risk assessment draft", "purpose": "Use when someone asks to draft the hazards, controls, PPE and residual risk for a permit." }, { "operation": "isolation_check", "name": "Isolation check", "purpose": "Use when someone asks whether the isolation for a job is actually in place." }, { "operation": "authorization_route", "name": "Sign-off route and timing", "purpose": "Use when someone asks where a permit is in the sign-off chain and whether it will be ready on time." }, { "operation": "crew_briefing", "name": "Crew briefing status", "purpose": "Use when someone asks whether the crew has acknowledged the safety briefing." }, { "operation": "clearance_check", "name": "Permit clearance check", "purpose": "Use when the crew has finished and someone asks whether the permit can be closed." }, { "operation": "safety_kpis", "name": "Permit safety roll-up", "purpose": "Use when a manager asks how permit safety is performing this month." }, { "operation": "work_request_check", "name": "Work request check", "purpose": "Use when someone asks whether a new work request is complete enough to raise a permit." }, { "operation": "isolation_confirmation", "name": "Control room isolation confirmation", "purpose": "Use when someone asks whether the control room has confirmed the isolation, or wants the permit update and crew notice drafted." }, { "operation": "acceptance_readiness", "name": "Permit acceptance readiness", "purpose": "Use when a crew lead asks whether the crew can accept the permit on site yet." }, { "operation": "audit_trail", "name": "Archived permit audit trail", "purpose": "Use when someone asks for the audit trail or chain of custody of an archived permit." } ], "easy_mode": [ "Verify Power Platform CLI 2.9.3 or newer and authenticate to the approved target environment.", "Initialize a modern cli-copilot project with an approved publisher prefix and preserve the generated identity settings.", "Upload the two packaged synthetic knowledge files and every operation-specific SKILL.md.", "Bind only approved least-privilege production connections after security, privacy, and business-owner review.", "Push the Draft agent, replay every locked persona case, and stop before publish for explicit approval." ], "local_install_prompt": "Install Field Permit to Work Agent from https://raw.githubusercontent.com/microsoft/aibast-agents-library/main/solutions/field-permit-to-work/deployment.json, verify the expected tool and smoke case, and preserve the synthetic read-only boundary. Do not ask me to open a terminal. Do not connect customer systems or execute external actions.", "copilot_studio_prompt": "Using the Microsoft Copilot Studio plugin, initialize a Draft agent for Field Permit to Work Agent from https://raw.githubusercontent.com/microsoft/aibast-agents-library/main/solutions/field-permit-to-work/deployment.json. Upload the two synthetic knowledge files and 11 skills, replay every locked case, and Stop before publish. Do not bind production connections without explicit customer approval.", "required_connections": [ "Use only approved, read-only, least-privilege connections to work-management, network control and crew briefing systems; the packaged demonstration uses synthetic knowledge files." ], "manual_commands": [ "pac auth create", "pac copilot init --name \"Field Permit to Work Agent\" --publisher-prefix --authoring-mode cli-copilot --project-dir \"\" --environment \"\"", "pac connection list --environment \"\"", "pac copilot pull --project-dir \"\"", "pac copilot push --project-dir \"\"", "pac copilot publish --bot \"\" --environment \"\"" ], "acceptance_checks": [ "Every exposed operation has one persona-language case and one SKILL.md.", "Exactly two synthetic knowledge files are present.", "Responses state the relevant human approval or review boundary.", "No operation claims an external write, communication, commitment, or protected decision.", "Publishing remains a separate user-approved step." ], "hard_mode": [ "Run pac copilot init and verify settings.mcs.yml before authoring components.", "Author global instructions that preserve the packaged synthetic-only and human-approval boundaries.", "Add the two knowledge files and one skill for every exposed operation.", "Create dedicated connection references only for approved systems and verify least-privilege access.", "Pull before push, inspect every generated file, replay every locked case, and stop before publish." ] }, "curated_copy": true, "source": "field", "origin": "both", "onepager": "60. Field Permit to Work Agent one-pager.pptx", "demo_video": "60-Field Permit to Work Agent.mp4", "has_onepager": true, "has_demo_video": true, "promise_coverage": 100, "is_primary": true, "aliases": [], "package": { "slug": "field-permit-to-work", "quest_url": "solutions/field-permit-to-work/quest.html", "manual_tutorial_url": "solutions/field-permit-to-work/manual-tutorial.html", "export_manifest_url": "solutions/field-permit-to-work/export-manifest.json" } }, "_demo": { "slug": "field-permit-to-work", "case_count": 11, "personas": [ "Authorized Person", "Crew Lead", "Operations Manager", "Permit Coordinator", "Safety Auditor" ], "prompts": [ "Which work permits are waiting for today?", "Draft the risk assessment for the Cedar Hill transformer job.", "Is the isolation for the Cedar Hill job actually in place?" ] }, "_catalog_kind": "solution", "_readiness": [ "Demo proven", "Conversation tested", "Synthetic data", "Production pattern" ] }, { "schema": "rapp-agent/1.0", "name": "@aibast-agents-library/field-service-dispatch", "version": "1.2.0", "display_name": "Field Service Dispatch Agent", "description": "Provide read-only field-service decision support from a synthetic metro-utility snapshot: tomorrow's job dashboard, schedule optimization with travel savings, skill-matched crew routes, an emergency outage response draft, live outage status with customer impact, the post-incident review, and prevention work orders with the monthly operations report. The agent never dispatches or reroutes a crew, changes a job, contacts a customer, stages inventory, creates a work order, or authorizes field action; a human dispatcher must approve work through future authenticated tools.", "author": "AIBAST", "tags": [ "field-service", "dispatch", "routing", "technicians", "emergency", "energy" ], "category": "energy", "quality_tier": "verified", "requires_env": [], "dependencies": [ "@rapp/basic-agent" ], "_file": "agents/@aibast-agents-library/energy_stacks/field_service_dispatch_stack/field_service_dispatch_agent.py", "_size_kb": 25.3, "_lines": 488, "_sha256": "49452ef7f2a119bb099f29c595124ea33414a3bd50623ef1812f02dd0fd1a846", "_install_prefix": "field_service_dispatch__", "_install_filename": "field_service_dispatch__49452ef7f2a1_agent.py", "_stack": "field_service_dispatch", "_stack_vertical": "energy", "_synthetic_data": true, "_added_at": "2026-03-18T13:23:12-04:00", "_solution": { "advertised_name": "Field Service Dispatch Agent", "sharepoint_list_name": "Field Service Dispatch Agent", "slot": 23, "executive_summary": "Automates field service scheduling and coordination to reduce response times, optimize resource use, and improve customer satisfaction.", "industries": [], "personas": [], "featured_tools": [ "Dynamics 365", "Microsoft Teams" ], "agent_requirements": [ "Microsoft Copilot Studio" ], "capabilities": [ "Detect outages via real-time IoT sensors, prioritizing critical facilities", "Dispatch nearest crews and reroute in-progress jobs automatically", "Provide real-time dashboards and automated customer notifications", "Optimize inventory staging and handle crew assignments" ], "outcomes": [ "Reduced emergency response times", "Improved first-time fix rates and eliminated redundant travel", "Delivered proactive customer updates, improving satisfaction", "Protected revenue and demonstrated operational excellence during crisis events" ], "customer_scenario": [ "Manual dispatch coordination caused delays and confusion", "Emergency response times exceeded SLA targets, risking public safety", "Managers lacked real-time visibility into crew locations and job progress", "Customers received slow updates, damaging trust" ], "slide_title": "Field Service Dispatch Agent", "slide_summary": "Automates field service scheduling and coordination to reduce response times, optimize resource use, and improve customer satisfaction.", "scenario_name": "A regional utility company", "challenge_intro": "A regional utility company struggled to effectively handle major outages:", "actions_intro": "The utility company leveraged the Field Service Dispatch Agent to:", "outcomes_intro": "With the agent\u2019s help, the utility company:", "opportunity_statements": [ "Perform intelligent route optimization and skill-based matching to reduce travel time and increase first-time fix rates", "Automate emergency response workflows with real-time crew tracking to protect revenue and public safety", "Deliver proactive customer communication and operational dashboards to boost customer confidence" ], "module_name": null, "module_description": null, "source_verified": true, "sales_headline": "Give dispatchers better options without letting the model move a crew", "card_pitch": "Turn synthetic service requests, zone capacity, certification evidence, and emergency context into dispatcher-reviewed options while keeping every job, route, message, inventory record, and field action unchanged.", "why_try": "A dispatch coordinator can ask who is eligible for a critical request and see the exact skill match and operational constraints before making a human decision.", "customer_challenge_copy": "Service requests, crew skills, current location, workload, and emergency priority are reviewed across disconnected systems, causing delay and confusion during high-pressure response.", "microsoft_ai_story": "Microsoft Copilot Studio can unify the conversation across Dynamics 365 Field Service, approved telematics, inventory context, Teams, and customer communication workflows after safety and authorization review.", "business_value": [ "Improves visibility into priority, capacity, and skill-match evidence.", "Supports more consistent dispatcher review of emergency and routine work.", "Protects field safety by keeping assignment, routing, notification, and execution in approved systems." ], "search_terms": [ "field service dispatch", "dispatch coordinator", "crew assignment", "route optimization", "emergency response", "technician certification", "Dynamics 365 Field Service", "telematics", "Microsoft Teams", "utilities", "energy", "Copilot Studio" ], "journey_stage": "Serve", "blueprint_role": "Creates a governed decision-support layer between field-service evidence and human dispatch authority, separate from the unadvertised grid-outage solution.", "sample_prompts": [ { "label": "Field Operations Manager: Job dashboard", "prompt": "What does tomorrow's job list look like by crew, and which crews are on standby?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=field-service-dispatch-dispatch-dashboard" }, { "label": "Service Director: Schedule optimization", "prompt": "I need help optimizing our field technician schedules for tomorrow. We have 15 service calls.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=field-service-dispatch-route-optimization" }, { "label": "Dispatch Coordinator: Crew route and skill match", "prompt": "Yes, show me Crew A-1's detailed route", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=field-service-dispatch-technician-assignment" }, { "label": "Emergency Duty Manager: Emergency response draft", "prompt": "URGENT: Major power outage downtown - 1,200 customers affected. Dispatch emergency crews now!", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=field-service-dispatch-emergency-response" }, { "label": "Control Room Operator: Live outage status", "prompt": "Yes, give me live updates and customer impact", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=field-service-dispatch-incident-status" }, { "label": "Reliability Engineer: Post-incident review", "prompt": "The outage is resolved. Show me the post-incident review and how we prevent the next one.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=field-service-dispatch-post-incident-review" }, { "label": "Operations Analyst: Work orders and monthly report", "prompt": "Yes, create the work orders and show me the monthly operations report", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=field-service-dispatch-work-orders-report" } ], "architecture": { "business_flow": [ "Field operations managers, dispatch coordinators, service directors, and emergency duty managers", "Microsoft 365 Copilot or Copilot Studio", "Field Service Dispatch Agent", "Dynamics 365 Field Service", "Approved telematics and inventory context", "Microsoft Teams and approved customer communication", "Human dispatcher and field supervisor" ], "capabilities": [ { "operation": "dispatch_dashboard", "name": "Job dashboard", "purpose": "Lists tomorrow's jobs by crew and crew status without changing a job or sending a message." }, { "operation": "route_optimization", "name": "Schedule optimization", "purpose": "Clusters tomorrow's jobs by geography and reports travel, skill match and cost savings as a draft schedule." }, { "operation": "technician_assignment", "name": "Crew route and skill match", "purpose": "Shows a crew's timed stops with the certification each stop needs, without assigning or notifying anyone." }, { "operation": "emergency_response", "name": "Emergency response draft", "purpose": "Prepares the outage dispatch plan, customer impact and schedule changes behind dispatcher approval." }, { "operation": "incident_status", "name": "Live outage status", "purpose": "Summarizes crew findings, customer impact per hour and the restoration timeline as a synthetic snapshot." }, { "operation": "post_incident_review", "name": "Post-incident review", "purpose": "Compares response against targets, computes value protected and ROI, and lists prevention actions." }, { "operation": "work_orders_report", "name": "Work orders and monthly report", "purpose": "Drafts the prevention work orders and summarizes the monthly operations metrics; nothing is created." } ], "easy_mode": [ "Verify Power Platform CLI 2.9.3 or newer and authenticate to the target environment.", "Use the Copilot Studio plugin initialization workflow to create a modern, sync-connected Draft agent with an approved publisher prefix.", "Upload manual/GLOBAL-INSTRUCTIONS.md, the two synthetic knowledge files, and all 4 operation skills from the solution package.", "Bind only approved, least-privilege production connections through dedicated connection references and preserve the human approval boundary.", "Push the Draft project, replay every locked persona prompt, inspect safety behavior, and stop before publishing." ], "local_install_prompt": "You are GitHub Copilot running in Agent mode in VS Code. Install and open the AI BAST Field Service Dispatch Agent from this deployment recipe: https://raw.githubusercontent.com/microsoft/aibast-agents-library/main/solutions/field-service-dispatch/deployment.json. Own the setup and do not ask me to open a terminal. Verify the Brainstem health endpoint, install or start it if needed using the official recipe, fetch the registry and source, verify the source hash, place the exact agent file in the discovered agents directory, confirm the expected tool, run the smoke prompt, and open the local UI. Preserve the synthetic read-only boundary. Do not connect customer systems, perform an external action, publish, or silently substitute another agent. Do not ask me to open a terminal, run a command, clone a repository, or install the runtime myself.", "copilot_studio_prompt": "You are GitHub Copilot running in Agent mode in VS Code with the Microsoft Copilot Studio plugin installed. Promote the AI BAST Field Service Dispatch Agent using this deployment recipe: https://raw.githubusercontent.com/microsoft/aibast-agents-library/main/solutions/field-service-dispatch/deployment.json. Ask only for the target environment ID and publisher prefix. Initialize a modern cli-copilot Draft project, upload the packaged global instructions, two synthetic knowledge files, and 4 skills, and preserve these operations: dispatch_dashboard, route_optimization, technician_assignment, emergency_response. Bind only approved dedicated connections, review every generated file and connection, pull before push, replay every locked persona case, enforce explicit authorized approval for any future write, and stop before publish. Show validation evidence and unresolved connection work. Stop before publish.", "required_connections": [ "Dynamics 365 Field Service", "Approved telematics and location feed", "Technician certification and labor records", "Parts and inventory availability", "Microsoft Teams and approved customer communication" ], "manual_commands": [ "pac auth create", "pac copilot init --name \"Field Service Dispatch Agent\" --publisher-prefix --authoring-mode cli-copilot --project-dir \"\" --environment \"\"", "pac connection list --environment \"\"", "pac copilot pull --project-dir \"\"", "pac copilot push --project-dir \"\"", "pac copilot publish --bot \"\" --environment \"\"" ], "acceptance_checks": [ "settings.mcs.yml preserves the initialized identity, environment, language, and authoring mode.", "All 4 implemented operations are represented by instructions, a packaged skill, or a bound tool.", "Exactly two synthetic knowledge files are loaded and clearly labeled fictional.", "Every production source uses a dedicated, least-privilege connection reference.", "Every locked persona prompt returns its deterministic synthetic evidence and the relevant human approval boundary.", "Never assign, notify, reroute, or dispatch a crew.", "Never change a job, stage inventory, contact a customer, or authorize field action.", "Public value statements remain qualitative; exact figures appear only as synthetic evidence.", "Publishing remains a separate, user-approved step after push and validation." ], "hard_mode": [ "Run pac copilot init and confirm settings.mcs.yml exists.", "Author global instructions that route every operation and preserve the packaged no-write boundaries.", "Upload the two synthetic Markdown knowledge files and 4 operation-specific SKILL.md files.", "Create read tools first; isolate any future write action behind authorization, current-state validation, explicit confirmation, and immutable audit logging.", "Validate every component schema and filename, push the Draft, replay every locked prompt, inspect failures, and stop before publish." ] }, "curated_copy": true, "source": "catalog", "origin": null, "onepager": "23. Field Service Dispatch Agent one-pager.pptx", "demo_video": null, "has_onepager": true, "has_demo_video": false, "promise_coverage": 60, "is_primary": true, "aliases": [], "package": { "slug": "field-service-dispatch", "quest_url": "solutions/field-service-dispatch/quest.html", "manual_tutorial_url": "solutions/field-service-dispatch/manual-tutorial.html", "export_manifest_url": "solutions/field-service-dispatch/export-manifest.json" } }, "_demo": { "slug": "field-service-dispatch", "case_count": 7, "personas": [ "Control Room Operator", "Dispatch Coordinator", "Emergency Duty Manager", "Field Operations Manager", "Operations Analyst", "Reliability Engineer", "Service Director" ], "prompts": [ "What does tomorrow's job list look like by crew, and which crews are on standby?", "I need help optimizing our field technician schedules for tomorrow. We have 15 service calls.", "Yes, show me Crew A-1's detailed route" ] }, "_catalog_kind": "solution", "_readiness": [ "Demo proven", "Conversation tested", "Synthetic data", "Production pattern" ] }, { "schema": "rapp-agent/1.0", "name": "@aibast-agents-library/grid-asset-maintenance", "version": "1.0.0", "display_name": "Predictive Asset Maintenance Intelligence Agent", "description": "Rank a distribution company's grid assets (power transformers, switchgear, underground cables and overhead lines) by failure probability over 30, 90 and 180 days from a synthetic condition-monitoring snapshot, draft the preventive work order with asset history, materials, long-lead parts and crew pre-assignment, check a completed job's field record before the asset register and fixed asset register are updated, and summarize fleet health with the replacement pipeline and capital needed. The agent never raises a work order, orders parts, assigns a crew or changes a register; an authorized engineer approves every step through future authenticated tools.", "author": "AIBAST", "tags": [ "energy", "utilities", "grid-assets", "predictive-maintenance", "asset-management", "reliability" ], "category": "energy", "quality_tier": "verified", "requires_env": [], "dependencies": [ "@rapp/basic-agent" ], "_file": "agents/@aibast-agents-library/energy_stacks/grid_asset_maintenance_stack/grid_asset_maintenance_agent.py", "_size_kb": 36.2, "_lines": 654, "_sha256": "02a7878dcf6124bbb90ef51592ee5c60851e2c043e2984e7e55fa8fea8c54cfd", "_install_prefix": "grid_asset_maintenance__", "_install_filename": "grid_asset_maintenance__02a7878dcf61_agent.py", "_stack": "grid_asset_maintenance", "_stack_vertical": "energy", "_synthetic_data": true, "_added_at": "2026-10-29T15:46:07-07:00", "_solution": { "advertised_name": "Predictive Asset Maintenance Intelligence Agent", "sharepoint_list_name": "Predictive Asset Maintenance Intelligence Agent", "slot": 72, "executive_summary": "Help an electricity distribution company maintain its grid by condition, not by calendar: rank transformers, switchgear, cables and overhead lines by 30, 90 and 180-day failure probability, draft preventive work orders with the parts and crew they need, check field records before the asset register changes, and show the replacement plan and capital by year.", "industries": [ "Energy and Utilities" ], "personas": [ "Reliability Engineer", "Maintenance Planner", "Asset Manager" ], "featured_tools": [ "Dynamics 365 Field Service", "Microsoft Dataverse", "Power BI" ], "agent_requirements": [ "Microsoft Copilot Studio" ], "capabilities": [ "Rank assets by 30, 90 and 180-day failure risk", "Draft work orders with history, parts and crew", "Hold register updates until the record is complete", "Show fleet health and replacement capital by year" ], "outcomes": [ "Sent crews to the riskiest assets first", "Ordered long-lead parts in time for the work", "Kept the asset register complete and current", "Built its replacement plan on condition data" ], "customer_scenario": [ "Substation condition data sat unused between cycles", "Long-lead parts were ordered after work was due", "Field records reached the register with tests missing", "Replacement plans rested on age, not condition" ], "slide_title": "Predictive Asset Maintenance Intelligence Agent", "slide_summary": "Ranks grid assets by failure risk, drafts the preventive work with its parts and crew, and keeps the asset register honest.", "scenario_name": "A regional distribution company", "challenge_intro": "A regional distribution company struggled to maintain its grid by condition:", "actions_intro": "The company leveraged the Predictive Asset Maintenance Intelligence Agent to:", "outcomes_intro": "With the agent\u2019s help, the company:", "opportunity_statements": [ "Rank transformers, switchgear, cables and lines by failure probability over 30, 90 and 180 days, with the anomaly behind each risk", "Draft preventive work orders with asset history, materials, long-lead parts to order first, and the crew and outage plan", "Keep the asset register complete by checking field records before updates, and show the replacement plan and capital by year" ], "module_name": null, "module_description": null, "source_verified": true, "sales_headline": "Find the grid assets most likely to fail and plan the work before they do", "card_pitch": "Give reliability engineers, maintenance planners and asset managers one conversational view of grid asset health: transformers, switchgear, cables and overhead lines ranked by 30, 90 and 180-day failure risk, preventive work orders drafted with the parts and crew they need, field records checked before the asset register changes, and the replacement plan with its capital.", "why_try": "The demo ranks a distribution company's grid assets by failure probability, drafts the work order for the riskiest transformer with the long-lead part to order first, holds a register update that is missing a test result, and shows the investment committee the replacement plan.", "customer_challenge_copy": "Distribution companies still maintain much of their grid on fixed cycles or after failures, so condition data sits unused, long-lead parts are ordered too late, field records reach the asset register incomplete, and replacement plans rest on age instead of condition.", "microsoft_ai_story": "Power Automate brings the condition-monitoring feed into Microsoft Dataverse, Copilot Studio ranks the risk and drafts the work in Microsoft Teams or Microsoft 365 Copilot, Dynamics 365 Field Service carries the approved work orders, Power Apps captures the field record, and Power BI shows fleet health and the replacement pipeline.", "business_value": [ "Ranks grid assets by 30, 90 and 180-day failure probability, so crews work on the riskiest assets first.", "Drafts preventive work orders with history, materials and the long-lead parts to order first.", "Holds asset register updates until the field record is complete, and shows the replacement capital by year." ], "search_terms": [ "Asset Manager", "Dynamics 365 Field Service", "Maintenance Planner", "Microsoft Copilot Studio", "Microsoft Dataverse", "Power Apps", "Power Automate", "Power BI", "Predictive Asset Maintenance Intelligence Agent", "Reliability Engineer", "asset health", "asset register", "condition-based maintenance", "distribution network", "failure probability", "grid assets", "overhead lines", "register_update", "replacement_plan", "risk_ranking", "switchgear", "transformers", "underground cables", "work_order_draft" ], "journey_stage": "Optimize", "blueprint_role": "Separates risk ranking and work preparation from protected work order, parts, crew, register and capital decisions.", "sample_prompts": [ { "label": "Reliability Engineer: Risk ranking", "prompt": "Which of our grid assets are most likely to fail in the next 90 days, and what's driving it?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=grid-asset-maintenance-risk-ranking" }, { "label": "Maintenance Planner: Work order draft", "prompt": "Get a preventive work order ready for the Millbrook transformer, with the parts and crew we need. Don't raise it yet.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=grid-asset-maintenance-work-order-draft" }, { "label": "Reliability Engineer: Register update check", "prompt": "The Mill Lane cable joint job closed yesterday. Is the field record complete enough to update the asset register?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=grid-asset-maintenance-register-update" }, { "label": "Director of Asset Management: Replacement plan", "prompt": "For the investment committee, how healthy is the fleet and what replacements and capital do we need over the next few years?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=grid-asset-maintenance-replacement-plan" } ], "architecture": { "business_flow": [ "Reliability engineers, maintenance planners and asset managers", "Microsoft Teams or Microsoft 365 Copilot", "Predictive Asset Maintenance Intelligence Agent", "Microsoft Dataverse asset health tables filled by Power Automate", "Dynamics 365 Field Service work orders and the Power Apps field job form", "Authorized approval by the planner, reliability engineer and investment committee" ], "capabilities": [ { "operation": "risk_ranking", "name": "Risk ranking", "purpose": "Use when a reliability engineer or planner asks which grid assets are most likely to fail in the next 30, 90 or 180 days, what is driving the risk, or what the maintenance backlog should be." }, { "operation": "work_order_draft", "name": "Work order draft", "purpose": "Use when a maintenance planner asks to prepare a preventive work order for an asset, with its history, materials, long-lead parts, crew or outage; returns a draft that is never raised." }, { "operation": "register_update", "name": "Register update check", "purpose": "Use when someone asks whether a completed maintenance job's field record is complete enough to update the asset register or fixed asset register." }, { "operation": "replacement_plan", "name": "Replacement plan", "purpose": "Use when an asset manager or director asks about fleet health by asset class, the replacement pipeline, or capital needed by year." } ], "easy_mode": [ "Verify Power Platform CLI 2.9.3 or newer and authenticate to the approved target environment.", "Initialize a modern cli-copilot project with an approved publisher prefix and preserve the generated identity settings.", "Upload the two packaged synthetic knowledge files and every operation-specific SKILL.md.", "Bind only approved least-privilege production connections after security, privacy, and business-owner review.", "Push the Draft agent, replay every locked persona case, and stop before publish for explicit approval." ], "local_install_prompt": "Install Predictive Asset Maintenance Intelligence Agent from https://raw.githubusercontent.com/microsoft/aibast-agents-library/main/solutions/grid-asset-maintenance/deployment.json, verify the expected tool and smoke case, and preserve the synthetic read-only boundary. Do not ask me to open a terminal. Do not connect utility systems or execute external actions.", "copilot_studio_prompt": "Using the Microsoft Copilot Studio plugin, initialize a Draft agent for Predictive Asset Maintenance Intelligence Agent from https://raw.githubusercontent.com/microsoft/aibast-agents-library/main/solutions/grid-asset-maintenance/deployment.json. Upload the two synthetic knowledge files and 4 skills, replay every locked case, and Stop before publish. Do not bind production connections without explicit customer approval.", "required_connections": [ "Use only approved, least-privilege production connections; the packaged demonstration uses synthetic knowledge files.", "Microsoft Dataverse asset health tables filled by Power Automate from the condition-monitoring system (read-only)", "Dynamics 365 Field Service for preventive work orders, raised only after a maintenance planner approves the draft", "ERP stores and fixed asset register through Power Automate, written only after a reliability engineer approves the update", "Microsoft Entra ID user identity for the approving planner and engineer" ], "manual_commands": [ "pac auth create", "pac copilot init --name \"Predictive Asset Maintenance Intelligence Agent\" --publisher-prefix --authoring-mode cli-copilot --project-dir \"\" --environment \"\"", "pac connection list --environment \"\"", "pac copilot pull --project-dir \"\"", "pac copilot push --project-dir \"\"", "pac copilot publish --bot \"\" --environment \"\"" ], "acceptance_checks": [ "Every exposed operation has one persona-language case and one SKILL.md.", "Exactly two synthetic knowledge files are present.", "Responses state the relevant human approval or review boundary.", "No operation claims a work order, part order, crew assignment, register change or capital decision.", "Publishing remains a separate user-approved step." ], "hard_mode": [ "Run pac copilot init and verify settings.mcs.yml before authoring components.", "Author global instructions that preserve the packaged synthetic-only and engineer-approval boundaries.", "Add the two knowledge files and one skill for every exposed operation.", "Create dedicated connection references only for approved systems and verify least-privilege access.", "Pull before push, inspect every generated file, replay every locked case, and stop before publish." ] }, "curated_copy": true, "source": "catalog", "origin": null, "onepager": "72. Predictive Asset Maintenance Intelligence Agent one-pager.pptx", "demo_video": null, "has_onepager": true, "has_demo_video": false, "promise_coverage": 100, "is_primary": true, "aliases": [], "package": { "slug": "grid-asset-maintenance", "quest_url": "solutions/grid-asset-maintenance/quest.html", "manual_tutorial_url": "solutions/grid-asset-maintenance/manual-tutorial.html", "export_manifest_url": "solutions/grid-asset-maintenance/export-manifest.json" } }, "_demo": { "slug": "grid-asset-maintenance", "case_count": 4, "personas": [ "Director of Asset Management", "Maintenance Planner", "Reliability Engineer" ], "prompts": [ "Which of our grid assets are most likely to fail in the next 90 days, and what's driving it?", "Get a preventive work order ready for the Millbrook transformer, with the parts and crew we need. Don't raise it yet.", "The Mill Lane cable joint job closed yesterday. Is the field record complete enough to update the asset register?" ] }, "_catalog_kind": "solution", "_readiness": [ "Demo proven", "Conversation tested", "Synthetic data", "Production pattern" ] }, { "schema": "rapp-agent/1.0", "name": "@aibast-agents-library/grid-outage-response", "version": "1.0.0", "display_name": "Grid Outage Response Agent", "description": "Coordinate outage detection, crew dispatch and customer communication to restore power faster, protect vulnerable customers, and defend regulatory reliability performance.", "author": "AIBAST", "tags": [ "energy", "utilities", "outage", "SAIDI", "SAIFI", "field-service", "reliability" ], "category": "energy", "quality_tier": "verified", "requires_env": [], "dependencies": [ "@rapp/basic-agent" ], "_file": "agents/@aibast-agents-library/energy_stacks/grid_outage_response_stack/grid_outage_response_agent.py", "_size_kb": 14.7, "_lines": 338, "_sha256": "198877c352c2715a2efa00bba06390cca79f4a2a88063cb2199f7f5cc69a4c47", "_install_prefix": "grid_outage_response__", "_install_filename": "grid_outage_response__198877c352c2_agent.py", "_stack": "grid_outage_response", "_stack_vertical": "energy", "_synthetic_data": true, "_added_at": "2026-08-07T18:32:19-04:00", "_solution": { "advertised_name": "Grid Outage Response Agent", "sharepoint_list_name": "Grid Outage Response Agent", "slot": null, "executive_summary": "Coordinate outage detection, crew dispatch and customer communication to restore power faster, protect vulnerable customers, and defend regulatory reliability performance.", "industries": [ "Energy and Utilities" ], "personas": [ "Control Room Operator", "Field Dispatch Supervisor", "Regulatory Reporting Manager" ], "featured_tools": [ "Dynamics 365 Field Service", "Azure IoT Hub", "Microsoft Teams", "Power BI" ], "agent_requirements": [], "capabilities": [ "Detect and rank active outages by customer impact and vulnerability", "Dispatch the right crew to the highest-impact circuit first", "Give every affected customer an accurate restoration estimate", "Track reliability minutes against the regulator's thresholds continuously" ], "outcomes": [ "Restored the highest-impact circuits first, cutting customer-minutes lost", "Protected critical-care customers with a register the control room could see", "Gave callers a restoration estimate that held, reducing repeat calls", "Tracked reliability performance continuously instead of reconstructing it at year end" ], "customer_scenario": [ "Storm outages were reported by customer calls before SCADA raised them", "Crews were dispatched without knowing which circuits carried critical-care customers", "Restoration estimates given to customers were guesses, driving repeat calls", "Reliability minutes were reconstructed manually at year end for the regulator" ], "slide_title": null, "slide_summary": null, "scenario_name": null, "challenge_intro": null, "actions_intro": null, "outcomes_intro": null, "opportunity_statements": [], "module_name": null, "module_description": null, "source_verified": false, "sales_headline": null, "card_pitch": null, "why_try": null, "customer_challenge_copy": null, "microsoft_ai_story": null, "business_value": [], "search_terms": [], "journey_stage": null, "blueprint_role": null, "sample_prompts": [], "architecture": null, "curated_copy": false, "source": null, "origin": null, "onepager": null, "demo_video": null, "has_onepager": false, "has_demo_video": false, "promise_coverage": null, "is_primary": true, "aliases": [], "package": { "slug": "grid-outage-response" } }, "_demo": { "slug": "grid-outage-response", "case_count": 4, "personas": [ "Control Room Operator", "Field Dispatch Supervisor", "Regulatory Reporting Manager" ], "prompts": [ "What's out right now and which one do I care about most?", "Who's rolling where, and is anything sitting with no crew on it?", "What are we telling customers about when their power comes back?" ] }, "_catalog_kind": "solution", "_readiness": [ "Demo proven", "Conversation tested", "Synthetic data", "Production pattern" ] }, { "schema": "rapp-agent/1.0", "name": "@aibast-agents-library/permit-license-management", "version": "1.2.0", "display_name": "Permit Management Agent", "description": "Review a synthetic renewables permit portfolio: status, permits expiring soon, renewal calendar, the emergency renewal package, compliance gaps, and application status. Use for permit and license tracking or audit-readiness questions. The agent never submits, renews, amends, or represents approval of a permit, and never sends notices; authorized staff must review and use future authenticated filing tools.", "author": "AIBAST", "tags": [ "permits", "licenses", "compliance", "regulatory", "energy", "renewals" ], "category": "energy", "quality_tier": "verified", "requires_env": [], "dependencies": [ "@rapp/basic-agent" ], "_file": "agents/@aibast-agents-library/energy_stacks/permit_license_management_stack/permit_license_management_agent.py", "_size_kb": 17.9, "_lines": 363, "_sha256": "40d35799eb84340dd98c5cca3ac896c99401688ebfe2660ab2f7b4ef53ab40e5", "_install_prefix": "permit_license_management__", "_install_filename": "permit_license_management__40d35799eb84_agent.py", "_stack": "permit_license_management", "_stack_vertical": "energy", "_synthetic_data": true, "_added_at": "2026-03-18T13:23:12-04:00", "_solution": { "advertised_name": "Permit Management Agent", "sharepoint_list_name": "Energy Operations Agent (b)", "slot": 11, "executive_summary": "Deliver real-time insights, automate critical workflows, and enable guided decision making\u2014boosting efficiency while reducing operational and compliance risk for energy organizations.", "industries": [ "Energy and Utilities" ], "personas": [ "Plant Manager / Reliability Engineer", "Compliance Manager", "Sustainability Lead", "Data Analyst" ], "featured_tools": [ "Dynamics 365", "SharePoint", "Microsoft Teams" ], "agent_requirements": [ "Microsoft Copilot Studio" ], "capabilities": [ "Predict failures with real-time analysis and automated planning", "Automate permit tracking and renewals with intelligent workflows", "Generate audit-ready reports quickly with streamlined data collection", "Consolidate fragmented data into real-time insights for easy centralized monitoring" ], "outcomes": [ "Improved asset reliability, reducing unplanned downtime and costs", "Maintained compliance and operational continuity across facilities", "Streamlined compliance reporting and minimized audit risk", "Gained real-time visibility and strategic planning support to achieve emissions goals" ], "customer_scenario": [ "Equipment failures interrupt operations and increase costs", "Hundreds of permits require heavy tracking and administrative work", "Compliance reporting can be slow, delaying audits and decisions", "Emissions data is hard to unify, making compliance tracking slow, error-prone, and difficult to manage" ], "slide_title": "Energy Operations Agent Suite", "slide_summary": "Deliver real-time insights, automate critical workflows, and enable guided decision making\u2014boosting efficiency while reducing operational and compliance risk for energy organizations.", "scenario_name": "A regional utility company", "challenge_intro": "A regional energy provider faced mounting challenges:", "actions_intro": "The energy provider deployed the Energy Operations Agent Suite to:", "outcomes_intro": "With the Agents, the energy provider:", "opportunity_statements": [], "module_name": "Permit Management Agent", "module_description": "Tracks permit status, automates renewal workflows, and maintains documentation to keep operations audit-ready", "source_verified": true, "sales_headline": "Keep permit evidence and renewal risk visible without bypassing authority", "card_pitch": "Turn a synthetic permit register, renewal calendar, compliance-gap view, and submitted-application tracker into a reviewable operating brief without renewing, amending, or filing anything.", "why_try": "Compliance and facility teams can ask which permit needs attention and receive the exact record, deadline, authority, and human review boundary.", "customer_challenge_copy": "Permit status, renewal lead times, supporting documents, and authority correspondence are fragmented, making it difficult to keep facilities audit-ready without heavy manual tracking.", "microsoft_ai_story": "Microsoft Copilot Studio can unify the conversation across an approved permit system, SharePoint controlled documents, Dynamics 365, Teams, and a separately governed authority-portal workflow.", "business_value": [ "Improves visibility into expiring and expired permit evidence.", "Creates a clearer handoff between facilities, compliance, and legal review.", "Protects filing authority by keeping renewals and submissions outside the analytical agent." ], "search_terms": [ "Energy Operations Agent Suite", "permit management", "license tracking", "renewal calendar", "compliance gaps", "application status", "environmental permits", "SharePoint controlled documents", "Dynamics 365", "Microsoft Teams", "Copilot Studio" ], "journey_stage": "Govern", "blueprint_role": "Preserves the Permit Management module as a distinct Energy Operations evidence layer before authorized legal and authority-portal action.", "sample_prompts": [ { "label": "Facility Manager: Permit status and expiring permits", "prompt": "Show me our permit status and any expiring soon", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=permit-license-management-permit-inventory" }, { "label": "Permit Coordinator: Renewal deadlines and lead times", "prompt": "Which permit renewals have already missed their lead time, and how many days are left?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=permit-license-management-renewal-calendar" }, { "label": "Compliance Manager: Compliance gap triage", "prompt": "What permit evidence gap needs immediate authorized review?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=permit-license-management-compliance-gaps" }, { "label": "Environmental Counsel: Application tracking", "prompt": "Where does the Imperial Valley Solar Phase 2 application stand, and did we submit anything today?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=permit-license-management-application-status" }, { "label": "Operations Director: Emergency renewal package", "prompt": "Start emergency renewals for the critical permits.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=permit-license-management-renewal-plan" } ], "architecture": { "business_flow": [ "Compliance managers, permit coordinators, facility managers, and environmental counsel", "Microsoft 365 Copilot or Copilot Studio", "Permit Management Agent", "Permit system of record", "SharePoint controlled documents", "Approved authority portal workflow", "Authorized permit owner and legal reviewer" ], "capabilities": [ { "operation": "permit_inventory", "name": "Permit status and expiring permits", "purpose": "Summarize the permit portfolio, permits expiring within 120 days, and the critical permits under 30 days." }, { "operation": "renewal_calendar", "name": "Renewal deadlines and lead times", "purpose": "List renewal deadlines with days left and lead-time status without initiating any renewal." }, { "operation": "compliance_gaps", "name": "Compliance gap triage", "purpose": "Triage missed renewal lead times and the obligations at risk without giving legal advice." }, { "operation": "application_status", "name": "Application tracking", "purpose": "Track already-submitted applications read-only." }, { "operation": "renewal_plan", "name": "Emergency renewal package", "purpose": "Prepare the emergency renewal package for the critical permits with costs, production protected, and draft stakeholder alerts." } ], "easy_mode": [ "Verify Power Platform CLI 2.9.3 or newer and authenticate to the target environment.", "Use the Copilot Studio plugin initialization workflow to create a modern, sync-connected Draft agent with an approved publisher prefix.", "Upload manual/GLOBAL-INSTRUCTIONS.md, the two synthetic knowledge files, and all 4 operation skills from the solution package.", "Bind only approved, least-privilege production connections through dedicated connection references and preserve the human approval boundary.", "Push the Draft project, replay every locked persona prompt, inspect safety behavior, and stop before publishing." ], "local_install_prompt": "You are GitHub Copilot running in Agent mode in VS Code. Install and open the AI BAST Permit Management Agent from this deployment recipe: https://raw.githubusercontent.com/microsoft/aibast-agents-library/main/solutions/permit-license-management/deployment.json. Own the setup and do not ask me to open a terminal. Verify the Brainstem health endpoint, install or start it if needed using the official recipe, fetch the registry and source, verify the source hash, place the exact agent file in the discovered agents directory, confirm the expected tool, run the smoke prompt, and open the local UI. Preserve the synthetic read-only boundary. Do not connect customer systems, perform an external action, publish, or silently substitute another agent. Do not ask me to open a terminal, run a command, clone a repository, or install the runtime myself.", "copilot_studio_prompt": "You are GitHub Copilot running in Agent mode in VS Code with the Microsoft Copilot Studio plugin installed. Promote the AI BAST Permit Management Agent using this deployment recipe: https://raw.githubusercontent.com/microsoft/aibast-agents-library/main/solutions/permit-license-management/deployment.json. Ask only for the target environment ID and publisher prefix. Initialize a modern cli-copilot Draft project, upload the packaged global instructions, two synthetic knowledge files, and 4 skills, and preserve these operations: permit_inventory, renewal_calendar, compliance_gaps, application_status. Bind only approved dedicated connections, review every generated file and connection, pull before push, replay every locked persona case, enforce explicit authorized approval for any future write, and stop before publish. Show validation evidence and unresolved connection work. Stop before publish.", "required_connections": [ "Permit system of record", "SharePoint controlled documents", "Dynamics 365 case management", "Approved authority portal connector", "Microsoft Teams for authorized review" ], "manual_commands": [ "pac auth create", "pac copilot init --name \"Permit Management Agent\" --publisher-prefix --authoring-mode cli-copilot --project-dir \"\" --environment \"\"", "pac connection list --environment \"\"", "pac copilot pull --project-dir \"\"", "pac copilot push --project-dir \"\"", "pac copilot publish --bot \"\" --environment \"\"" ], "acceptance_checks": [ "settings.mcs.yml preserves the initialized identity, environment, language, and authoring mode.", "All 4 implemented operations are represented by instructions, a packaged skill, or a bound tool.", "Exactly two synthetic knowledge files are loaded and clearly labeled fictional.", "Every production source uses a dedicated, least-privilege connection reference.", "Every locked persona prompt returns its deterministic synthetic evidence and the relevant human approval boundary.", "Never submit, renew, amend, withdraw, or represent approval of a permit.", "Treat compliance findings as review evidence, not legal advice.", "Public value statements remain qualitative; exact figures appear only as synthetic evidence.", "Publishing remains a separate, user-approved step after push and validation." ], "hard_mode": [ "Run pac copilot init and confirm settings.mcs.yml exists.", "Author global instructions that route every operation and preserve the packaged no-write boundaries.", "Upload the two synthetic Markdown knowledge files and 4 operation-specific SKILL.md files.", "Create read tools first; isolate any future write action behind authorization, current-state validation, explicit confirmation, and immutable audit logging.", "Validate every component schema and filename, push the Draft, replay every locked prompt, inspect failures, and stop before publish." ] }, "curated_copy": true, "source": "catalog", "origin": null, "onepager": "Energy Operations Agent Suite one-pager (2).pptx", "demo_video": "11-14-Energy Operations Suite.mp4", "has_onepager": true, "has_demo_video": true, "promise_coverage": 85, "is_primary": true, "aliases": [], "package": { "slug": "permit-license-management", "quest_url": "solutions/permit-license-management/quest.html", "manual_tutorial_url": "solutions/permit-license-management/manual-tutorial.html", "export_manifest_url": "solutions/permit-license-management/export-manifest.json" } }, "_demo": { "slug": "permit-license-management", "case_count": 5, "personas": [ "Compliance Manager", "Environmental Counsel", "Facility Manager", "Operations Director", "Permit Coordinator" ], "prompts": [ "Show me our permit status and any expiring soon", "Which permit renewals have already missed their lead time, and how many days are left?", "What permit evidence gap needs immediate authorized review?" ] }, "_catalog_kind": "solution", "_readiness": [ "Demo proven", "Conversation tested", "Synthetic data", "Production pattern" ] }, { "schema": "rapp-agent/1.0", "name": "@aibast-agents-library/procurement-supplier-collaboration", "version": "1.0.0", "display_name": "Procurement and Supplier Collaboration Agent", "description": "Run a utility's source-to-pay cycle for network materials from a synthetic snapshot: requisitions from the maintenance plan, capital projects and stock replenishment with the recommended route (framework direct award, mini-competition or open tender) and eligible supplier shortlist, answers to supplier questions from the published RFQ pack, weighted bid scoring with a ranked recommendation, purchase order acknowledgement and ASN delivery deviations, and the three-way match of purchase order, goods receipt and invoice. The agent never awards, orders, messages a supplier or approves a payment; an authorized person approves every step through future authenticated tools.", "author": "AIBAST", "tags": [ "procurement", "sourcing", "supplier-collaboration", "bid-evaluation", "three-way-match", "energy", "utilities" ], "category": "energy", "quality_tier": "verified", "requires_env": [], "dependencies": [ "@rapp/basic-agent" ], "_file": "agents/@aibast-agents-library/energy_stacks/procurement_supplier_collaboration_stack/procurement_supplier_collaboration_agent.py", "_size_kb": 38.6, "_lines": 647, "_sha256": "4c87b37ed142c86aa8350f49c3947987f327b82ed07692e0e5f4952a893d2524", "_install_prefix": "procurement_supplier_collaboration__", "_install_filename": "procurement_supplier_collaboration__4c87b37ed142_agent.py", "_stack": "procurement_supplier_collaboration", "_stack_vertical": "energy", "_synthetic_data": true, "_added_at": "2026-10-29T17:55:05-07:00", "_solution": { "advertised_name": "Procurement and Supplier Collaboration Agent", "sharepoint_list_name": "Procurement and Supplier Collaboration Agent", "slot": 73, "executive_summary": "Help a utility buy critical network materials faster: recommend the procurement route and supplier shortlist for every requisition, answer supplier questions from the RFQ pack, score bids on price, lead time, quality and sustainability, flag late deliveries, and hold invoices that fail the three-way match, while every award and payment stays with an authorized person.", "industries": [ "Energy and Utilities" ], "personas": [ "Category Manager", "Buyer", "Accounts Payable Lead" ], "featured_tools": [ "Microsoft Dataverse", "Power Apps", "Power Automate" ], "agent_requirements": [ "Microsoft Copilot Studio" ], "capabilities": [ "Recommend the route and supplier shortlist", "Answer bidders from the published RFQ pack", "Score bids on price, lead time and quality", "Flag late deliveries and unmatched invoices" ], "outcomes": [ "Shortened the cycle for routine orders", "Treated every bidder the same way", "Gave the award panel a clear recommendation", "Paid only invoices that matched the order" ], "customer_scenario": [ "Procurement routes differed from buyer to buyer", "Supplier questions sat unanswered in email", "Bid evaluations were hard to explain", "Late shipments and invoice errors surfaced late" ], "slide_title": "Procurement and Supplier Collaboration Agent", "slide_summary": "Recommends how to buy, scores bids fairly, chases late deliveries and holds invoices that do not match.", "scenario_name": "A regional distribution company", "challenge_intro": "A regional distribution company struggled to buy network materials on time:", "actions_intro": "The company leveraged the Procurement and Supplier Collaboration Agent to:", "outcomes_intro": "With the agent\u2019s help, the company:", "opportunity_statements": [ "Recommend the procurement route and an eligible supplier shortlist for requisitions from maintenance, projects and stock", "Answer suppliers and score bids from the published RFQ pack, ranking price, lead time, quality and sustainability", "Chase deliveries and match invoices, flagging late shipping notices and holding invoices that fail the three-way match" ], "module_name": null, "module_description": null, "source_verified": true, "sales_headline": "Buy network materials faster, from requisition to matched invoice", "card_pitch": "Give category managers, buyers, expediters and accounts payable one conversational view of source-to-pay for network materials: the right procurement route and supplier shortlist, supplier questions answered from the RFQ pack, bids scored and explained, late deliveries flagged, and invoices matched before payment.", "why_try": "The demo recommends the route and shortlist for a transformer order, drafts a fair answer to a bidder's question, ranks three bids for the award panel, flags a cable delivery that puts a project at risk, and holds two invoices that fail the three-way match.", "customer_challenge_copy": "Utilities buy critical network materials through slow, manual cycles: routes are chosen inconsistently, supplier questions sit in email, bid evaluations are hard to explain, late shipments surface when crews are already waiting, and invoice mismatches are found after payment.", "microsoft_ai_story": "Power Automate brings ERP requisitions, orders, receipts and invoices into Microsoft Dataverse, suppliers work in a Power Apps supplier portal, and Copilot Studio recommends routes, answers supplier questions, scores bids and checks deliveries and invoices in Microsoft Teams or Microsoft 365 Copilot, with every award and payment approved by a person.", "business_value": [ "Recommends the procurement route and an eligible supplier shortlist for every requisition.", "Scores bids on price, lead time, quality and sustainability with a narrative the award panel can defend.", "Flags late deliveries before crews wait, and holds invoices that fail the three-way match." ], "search_terms": [ "Accounts Payable", "Buyer", "Category Manager", "Expediter", "Microsoft Copilot Studio", "Microsoft Dataverse", "Power Apps", "Power Automate", "Procurement and Supplier Collaboration Agent", "advance shipping notice", "bid evaluation", "bid_evaluation", "delivery_tracking", "framework agreement", "invoice_match", "network materials", "purchase order", "request for quotation", "sourcing", "sourcing_plan", "supplier portal", "supplier_query", "three-way match" ], "journey_stage": "Optimize", "blueprint_role": "Separates sourcing recommendations, bid scoring and match checks from protected awards, orders, supplier messages and payments.", "sample_prompts": [ { "label": "Category Manager: Sourcing plan", "prompt": "What requisitions are waiting, how should we buy each one, and who should we invite for the transformer order?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=procurement-supplier-collaboration-sourcing-plan" }, { "label": "Buyer: Supplier question", "prompt": "A bidder on the transformer RFQ asks if they can deliver all 24 units in two lots instead of three. What should I tell them?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=procurement-supplier-collaboration-supplier-query" }, { "label": "Category Manager: Bid evaluation", "prompt": "The transformer bids are in. How do they score, and who should the award panel pick?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=procurement-supplier-collaboration-bid-evaluation" }, { "label": "Expediter: Delivery tracking", "prompt": "Which of our open orders have delivery problems I need to chase today?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=procurement-supplier-collaboration-delivery-tracking" }, { "label": "Accounts Payable Lead: Three-way match", "prompt": "Which supplier invoices can go for payment approval today, and which ones need to be held?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=procurement-supplier-collaboration-invoice-match" } ], "architecture": { "business_flow": [ "Category managers, buyers, expediters and accounts payable", "Microsoft Teams or Microsoft 365 Copilot", "Procurement and Supplier Collaboration Agent", "Microsoft Dataverse tables filled by Power Automate from the ERP", "Power Apps supplier portal: RFQs, bids, acknowledgements and shipping notices", "Authorized approval by the category manager, award panel and accounts payable" ], "capabilities": [ { "operation": "sourcing_plan", "name": "Sourcing plan", "purpose": "Use when a category manager or buyer asks which requisitions are open, how each should be bought, or which suppliers to invite." }, { "operation": "supplier_query", "name": "Supplier question", "purpose": "Use when a buyer asks what to tell a supplier who has a question about an open RFQ; returns a draft reply that is never sent." }, { "operation": "bid_evaluation", "name": "Bid evaluation", "purpose": "Use when someone asks how the bids score on price, lead time, quality and sustainability, or which supplier should win." }, { "operation": "delivery_tracking", "name": "Delivery tracking", "purpose": "Use when an expediter or buyer asks which purchase orders are unacknowledged or late against the committed date." }, { "operation": "invoice_match", "name": "Three-way match", "purpose": "Use when accounts payable asks which invoices pass the three-way match for payment approval and which must be held." } ], "easy_mode": [ "Verify Power Platform CLI 2.9.3 or newer and authenticate to the approved target environment.", "Initialize a modern cli-copilot project with an approved publisher prefix and preserve the generated identity settings.", "Upload the two packaged synthetic knowledge files and every operation-specific SKILL.md.", "Bind only approved least-privilege production connections after security, privacy, and business-owner review.", "Push the Draft agent, replay every locked persona case, and stop before publish for explicit approval." ], "local_install_prompt": "Install Procurement and Supplier Collaboration Agent from https://raw.githubusercontent.com/microsoft/aibast-agents-library/main/solutions/procurement-supplier-collaboration/deployment.json, verify the expected tool and smoke case, and preserve the synthetic read-only boundary. Do not ask me to open a terminal. Do not connect procurement or finance systems or execute external actions.", "copilot_studio_prompt": "Using the Microsoft Copilot Studio plugin, initialize a Draft agent for Procurement and Supplier Collaboration Agent from https://raw.githubusercontent.com/microsoft/aibast-agents-library/main/solutions/procurement-supplier-collaboration/deployment.json. Upload the two synthetic knowledge files and 5 skills, replay every locked case, and Stop before publish. Do not bind production connections without explicit customer approval.", "required_connections": [ "Use only approved, least-privilege production connections; the packaged demonstration uses synthetic knowledge files.", "Microsoft Dataverse tables filled by Power Automate from the ERP: requisitions, frameworks, purchase orders, goods receipts and invoices (read-only)", "Power Apps supplier portal for RFQs, bids, clarifications, acknowledgements and advance shipping notices, with authenticated supplier access", "ERP finance module for payment approval, used only by accounts payable after the three-way match", "Microsoft Entra ID user identity for buyers, category managers and accounts payable approvers" ], "manual_commands": [ "pac auth create", "pac copilot init --name \"Procurement and Supplier Collaboration Agent\" --publisher-prefix --authoring-mode cli-copilot --project-dir \"\" --environment \"\"", "pac connection list --environment \"\"", "pac copilot pull --project-dir \"\"", "pac copilot push --project-dir \"\"", "pac copilot publish --bot \"\" --environment \"\"" ], "acceptance_checks": [ "Every exposed operation has one persona-language case and one SKILL.md.", "Exactly two synthetic knowledge files are present.", "Responses state the relevant human approval or review boundary.", "No operation claims an RFQ, award, order, supplier message, goods receipt or payment.", "Publishing remains a separate user-approved step." ], "hard_mode": [ "Run pac copilot init and verify settings.mcs.yml before authoring components.", "Author global instructions that preserve the packaged synthetic-only, equal-treatment and approval boundaries.", "Add the two knowledge files and one skill for every exposed operation.", "Create dedicated connection references only for approved systems and verify least-privilege access.", "Pull before push, inspect every generated file, replay every locked case, and stop before publish." ] }, "curated_copy": true, "source": "catalog", "origin": null, "onepager": "73. Procurement and Supplier Collaboration Agent one-pager.pptx", "demo_video": null, "has_onepager": true, "has_demo_video": false, "promise_coverage": 100, "is_primary": true, "aliases": [], "package": { "slug": "procurement-supplier-collaboration", "quest_url": "solutions/procurement-supplier-collaboration/quest.html", "manual_tutorial_url": "solutions/procurement-supplier-collaboration/manual-tutorial.html", "export_manifest_url": "solutions/procurement-supplier-collaboration/export-manifest.json" } }, "_demo": { "slug": "procurement-supplier-collaboration", "case_count": 5, "personas": [ "Accounts Payable Lead", "Buyer", "Category Manager", "Expediter" ], "prompts": [ "What requisitions are waiting, how should we buy each one, and who should we invite for the transformer order?", "A bidder on the transformer RFQ asks if they can deliver all 24 units in two lots instead of three. What should I tell them?", "The transformer bids are in. How do they score, and who should the award panel pick?" ] }, "_catalog_kind": "solution", "_readiness": [ "Demo proven", "Conversation tested", "Synthetic data", "Production pattern" ] }, { "schema": "rapp-agent/1.0", "name": "@aibast-agents-library/energy-regulatory-reporting", "version": "1.1.0", "display_name": "Regulatory Reporting Agent", "description": "Review synthetic energy regulatory-report status, data-validation exceptions, submission tracking, and audit-readiness findings. Use for EPA, FERC, PHMSA, or state reporting preparation. The agent never files, certifies, signs, or transmits a report; an authorized regulatory owner must approve a source-backed package through future authenticated tools.", "author": "AIBAST", "tags": [ "regulatory", "reporting", "epa", "ferc", "audit", "compliance", "energy" ], "category": "energy", "quality_tier": "verified", "requires_env": [], "dependencies": [ "@rapp/basic-agent" ], "_file": "agents/@aibast-agents-library/energy_stacks/regulatory_reporting_stack/regulatory_reporting_agent.py", "_size_kb": 21.5, "_lines": 497, "_sha256": "c0045a249275152acee2683de879c34a478c52544868e0a592a7f7536c05a3e5", "_install_prefix": "energy_regulatory_reporting__", "_install_filename": "energy_regulatory_reporting__c0045a249275_agent.py", "_stack": "regulatory_reporting", "_stack_vertical": "energy", "_synthetic_data": true, "_added_at": "2026-03-18T13:23:12-04:00", "_solution": { "advertised_name": "Regulatory Reporting Agent", "sharepoint_list_name": "Energy Operations Agent (c)", "slot": 17, "executive_summary": "Deliver real-time insights, automate critical workflows, and enable guided decision making\u2014boosting efficiency while reducing operational and compliance risk for energy organizations.", "industries": [ "Energy and Utilities" ], "personas": [ "Plant Manager / Reliability Engineer", "Compliance Manager", "Sustainability Lead", "Data Analyst" ], "featured_tools": [ "Dynamics 365", "SharePoint", "Microsoft Teams" ], "agent_requirements": [ "Microsoft Copilot Studio" ], "capabilities": [ "Predict failures with real-time analysis and automated planning", "Automate permit tracking and renewals with intelligent workflows", "Generate audit-ready reports quickly with streamlined data collection", "Consolidate fragmented data into real-time insights for easy centralized monitoring" ], "outcomes": [ "Improved asset reliability, reducing unplanned downtime and costs", "Maintained compliance and operational continuity across facilities", "Streamlined compliance reporting and minimized audit risk", "Gained real-time visibility and strategic planning support to achieve emissions goals" ], "customer_scenario": [ "Equipment failures interrupt operations and increase costs", "Hundreds of permits require heavy tracking and administrative work", "Compliance reporting can be slow, delaying audits and decisions", "Emissions data is hard to unify, making compliance tracking slow, error-prone, and difficult to manage" ], "slide_title": "Energy Operations Agent Suite", "slide_summary": "Deliver real-time insights, automate critical workflows, and enable guided decision making\u2014boosting efficiency while reducing operational and compliance risk for energy organizations.", "scenario_name": "A regional utility company", "challenge_intro": "A regional energy provider faced mounting challenges:", "actions_intro": "The energy provider deployed the Energy Operations Agent Suite to:", "outcomes_intro": "With the Agents, the energy provider:", "opportunity_statements": [], "module_name": "Regulatory Reporting Agent", "module_description": "Consolidates data, validates compliance requirements, and generates regulatory reports to streamline submissions", "source_verified": true, "sales_headline": "Make reporting gaps reviewable before an authorized filing", "card_pitch": "Bring synthetic report status, source-quality exceptions, filing-state evidence, and audit findings into one regulatory brief while leaving certification, signature, and transmission to authorized owners.", "why_try": "A reporting lead can ask what is overdue or incomplete and receive the exact report, evidence gap, owner, and approval boundary.", "customer_challenge_copy": "Regulatory data, completion status, validation evidence, and audit findings live across systems, slowing preparation and obscuring who must resolve each issue.", "microsoft_ai_story": "Microsoft Copilot Studio can provide the governed front door; approved regulatory data, a controlled reporting workspace, SharePoint, Teams, and a separately authorized filing connector can support production use.", "business_value": [ "Improves visibility into overdue reports and source-quality exceptions.", "Creates a consistent audit-readiness handoff with named evidence and owners.", "Preserves regulatory accountability by separating preparation from certification and filing." ], "search_terms": [ "Energy Operations Agent Suite", "energy regulatory reporting", "EPA reporting", "FERC Form 1", "PHMSA reporting", "data validation", "audit readiness", "regulator filing", "SharePoint", "Microsoft Teams", "Copilot Studio" ], "journey_stage": "Govern", "blueprint_role": "Preserves the Regulatory Reporting module as a distinct Energy Operations preparation layer before certification, signature, and filing.", "sample_prompts": [ { "label": "Regulatory Reporting Lead: Reporting portfolio status", "prompt": "Which filing is overdue and who owns it?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=energy-regulatory-reporting-report-status" }, { "label": "Data Analyst: Source-data validation screening", "prompt": "Which report data is incomplete or below quality threshold?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=energy-regulatory-reporting-data-validation" }, { "label": "Compliance Manager: Read-only filing-state tracker", "prompt": "Show filing state and confirm you did not transmit anything.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=energy-regulatory-reporting-submission-tracker" }, { "label": "Internal Auditor: Audit evidence triage", "prompt": "What high-severity reporting evidence is still open?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=energy-regulatory-reporting-audit-readiness" }, { "label": "Environmental Manager: Quarterly emissions summary", "prompt": "I need to prepare our quarterly EPA emissions report. Can you pull the data?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=energy-regulatory-reporting-emissions-summary" }, { "label": "Regulatory Affairs Director: Draft EPA submission package", "prompt": "Generate the EPA submission file for the quarter and show me any compliance risks.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=energy-regulatory-reporting-prepare-submission" } ], "architecture": { "business_flow": [ "Compliance managers, regulatory reporting leads, data analysts, and internal auditors", "Microsoft 365 Copilot or Copilot Studio", "Regulatory Reporting Agent", "Approved regulatory data lake", "Controlled reporting workspace", "Approved filing connector and Microsoft Teams", "Authorized regulatory owner" ], "capabilities": [ { "operation": "report_status", "name": "Reporting portfolio status", "purpose": "Summarizes synthetic deadlines, completion, quality, filing state, and ownership." }, { "operation": "data_validation", "name": "Source-data validation screening", "purpose": "Identifies incomplete or below-threshold synthetic source evidence before certification." }, { "operation": "submission_tracker", "name": "Read-only filing-state tracker", "purpose": "Shows recorded submission state without signing, certifying, or transmitting a filing." }, { "operation": "audit_readiness", "name": "Audit evidence triage", "purpose": "Organizes open findings and severity without predicting an audit outcome or providing legal advice." }, { "operation": "emissions_summary", "name": "Quarterly emissions summary", "purpose": "Consolidates synthetic quarterly pollutant totals against annual permit limits across all facilities." }, { "operation": "prepare_submission", "name": "Draft EPA submission package", "purpose": "Prepares a draft CAMD submission file and lists compliance risks without uploading or filing anything." } ], "easy_mode": [ "Verify Power Platform CLI 2.9.3 or newer and authenticate to the target environment.", "Use the Copilot Studio plugin initialization workflow to create a modern, sync-connected Draft agent with an approved publisher prefix.", "Upload manual/GLOBAL-INSTRUCTIONS.md, the two synthetic knowledge files, and all 4 operation skills from the solution package.", "Bind only approved, least-privilege production connections through dedicated connection references and preserve the human approval boundary.", "Push the Draft project, replay every locked persona prompt, inspect safety behavior, and stop before publishing." ], "local_install_prompt": "You are GitHub Copilot running in Agent mode in VS Code. Install and open the AI BAST Regulatory Reporting Agent from this deployment recipe: https://raw.githubusercontent.com/microsoft/aibast-agents-library/main/solutions/energy-regulatory-reporting/deployment.json. Own the setup and do not ask me to open a terminal. Verify the Brainstem health endpoint, install or start it if needed using the official recipe, fetch the registry and source, verify the source hash, place the exact agent file in the discovered agents directory, confirm the expected tool, run the smoke prompt, and open the local UI. Preserve the synthetic read-only boundary. Do not connect customer systems, perform an external action, publish, or silently substitute another agent. Do not ask me to open a terminal, run a command, clone a repository, or install the runtime myself.", "copilot_studio_prompt": "You are GitHub Copilot running in Agent mode in VS Code with the Microsoft Copilot Studio plugin installed. Promote the AI BAST Regulatory Reporting Agent using this deployment recipe: https://raw.githubusercontent.com/microsoft/aibast-agents-library/main/solutions/energy-regulatory-reporting/deployment.json. Ask only for the target environment ID and publisher prefix. Initialize a modern cli-copilot Draft project, upload the packaged global instructions, two synthetic knowledge files, and 4 skills, and preserve these operations: report_status, data_validation, submission_tracker, audit_readiness. Bind only approved dedicated connections, review every generated file and connection, pull before push, replay every locked persona case, enforce explicit authorized approval for any future write, and stop before publish. Show validation evidence and unresolved connection work. Stop before publish.", "required_connections": [ "Approved regulatory data lake", "Controlled reporting workspace", "SharePoint evidence library", "Approved regulator filing connector", "Microsoft Teams for review and escalation" ], "manual_commands": [ "pac auth create", "pac copilot init --name \"Regulatory Reporting Agent\" --publisher-prefix --authoring-mode cli-copilot --project-dir \"\" --environment \"\"", "pac connection list --environment \"\"", "pac copilot pull --project-dir \"\"", "pac copilot push --project-dir \"\"", "pac copilot publish --bot \"\" --environment \"\"" ], "acceptance_checks": [ "settings.mcs.yml preserves the initialized identity, environment, language, and authoring mode.", "All 4 implemented operations are represented by instructions, a packaged skill, or a bound tool.", "Exactly two synthetic knowledge files are loaded and clearly labeled fictional.", "Every production source uses a dedicated, least-privilege connection reference.", "Every locked persona prompt returns its deterministic synthetic evidence and the relevant human approval boundary.", "Never sign, certify, transmit, or claim completion of a regulator filing.", "Never predict an audit outcome or provide legal advice.", "Public value statements remain qualitative; exact figures appear only as synthetic evidence.", "Publishing remains a separate, user-approved step after push and validation." ], "hard_mode": [ "Run pac copilot init and confirm settings.mcs.yml exists.", "Author global instructions that route every operation and preserve the packaged no-write boundaries.", "Upload the two synthetic Markdown knowledge files and 4 operation-specific SKILL.md files.", "Create read tools first; isolate any future write action behind authorization, current-state validation, explicit confirmation, and immutable audit logging.", "Validate every component schema and filename, push the Draft, replay every locked prompt, inspect failures, and stop before publish." ] }, "curated_copy": true, "source": "catalog", "origin": null, "onepager": "Energy Operations Agent Suite one-pager (2).pptx", "demo_video": "11-14-Energy Operations Suite.mp4", "has_onepager": true, "has_demo_video": true, "promise_coverage": 85, "is_primary": true, "aliases": [], "package": { "slug": "energy-regulatory-reporting", "quest_url": "solutions/energy-regulatory-reporting/quest.html", "manual_tutorial_url": "solutions/energy-regulatory-reporting/manual-tutorial.html", "export_manifest_url": "solutions/energy-regulatory-reporting/export-manifest.json" } }, "_demo": { "slug": "energy-regulatory-reporting", "case_count": 6, "personas": [ "Compliance Manager", "Data Analyst", "Environmental Manager", "Internal Auditor", "Regulatory Affairs Director", "Regulatory Reporting Lead" ], "prompts": [ "Which filing is overdue and who owns it?", "Which report data is incomplete or below quality threshold?", "Show filing state and confirm you did not transmit anything." ] }, "_catalog_kind": "solution", "_readiness": [ "Demo proven", "Conversation tested", "Synthetic data", "Production pattern" ] }, { "schema": "rapp-agent/1.0", "name": "@aibast-agents-library/claims-processing", "version": "1.0.0", "display_name": "Claims Processing Agent", "description": "Automate claims processing workflows to deliver faster, consistent, and more compliant claim outcomes.", "author": "AIBAST", "tags": [ "claims", "insurance", "adjudication", "fraud", "settlement", "financial-services" ], "category": "financial_services", "quality_tier": "verified", "requires_env": [], "dependencies": [ "@rapp/basic-agent" ], "_file": "agents/@aibast-agents-library/financial_services_stacks/claims_processing_stack/claims_processing_agent.py", "_size_kb": 33.9, "_lines": 662, "_sha256": "81dca988b558093374ea92d739d2a4aa9ba6f5aa8da568f77ac006a6e6f31e1e", "_install_prefix": "claims_processing__", "_install_filename": "claims_processing__81dca988b558_agent.py", "_stack": "claims_processing", "_stack_vertical": "financial_services", "_synthetic_data": true, "_added_at": "2026-03-18T13:23:12-04:00", "_solution": { "advertised_name": "Claims Processing Agent", "sharepoint_list_name": "Claims Processing Agent", "slot": 36, "executive_summary": "Automate claims processing workflows to deliver faster, consistent, and more compliant claim outcomes.", "industries": [ "Financial Services" ], "personas": [ "Claims Adjusters / Managers", "SIU Teams", "Operations Leaders" ], "featured_tools": [ "Dynamics 365", "Microsoft Teams" ], "agent_requirements": [ "Microsoft Copilot Studio" ], "capabilities": [ "Classify and route claims instantly to automated or human workflows", "Flag suspected fraud early and generate SIU evidence packages", "Pre\u2011analyze complex claims and highlight missing documents", "Auto\u2011generate approval or denial recommendations aligned with guidelines" ], "outcomes": [ "Cut processing time dramatically with automated workflows", "Reduced fraud exposure with earlier detection and SIU referrals", "Accelerated adjudication with ready\u2011for\u2011review claim files", "Improved service quality with faster decisions and communications", "Increased operational capacity" ], "customer_scenario": [ "Incoming claims lacked standardized triage and consumed adjuster time", "Fraud indicators were often detected late in the process", "Simple claims were processed inefficiently alongside complex ones", "Strict compliance deadlines and documentation requirements presented unmitigated risks" ], "slide_title": "Claims Processing Agent", "slide_summary": "Automates claims processing workflows to deliver faster, consistent, and more compliant claim outcomes.", "scenario_name": "An insurance company", "challenge_intro": "An insurance company struggled with high claim volumes and long cycle times:", "actions_intro": "The insurance company used the Claims Processing Agent to:", "outcomes_intro": "With the agent in place, the company:", "opportunity_statements": [ "Triage claims instantly to accelerate routing and reduce adjuster workloads", "Spot fraud indicators early to prevent exposure and protect claim payouts", "Auto\u2011adjudicate simple claims to shorten cycle times and improve consistency", "Pre\u2011prepare complex claim files to give adjusters ready\u2011to\u2011review summaries" ], "module_name": null, "module_description": null, "source_verified": true, "sales_headline": "Prepare every claim for the right human review", "card_pitch": "Bring intake, policy terms, documents, fraud indicators, and nonbinding settlement estimates into one adjuster workspace without claiming approval, denial, payment, or coverage action.", "why_try": "A claims manager can ask which file needs attention and see the exact synthetic claim, missing evidence, SIU indicator, policy term, and next review checkpoint.", "customer_challenge_copy": "An insurer is processing claims through separate intake, policy, document, SIU, and adjudication systems. Adjusters spend time rebuilding files while automated scoring can be mistaken for a coverage or fraud decision.", "microsoft_ai_story": "Microsoft Copilot Studio provides the adjuster-facing experience. Dynamics 365 manages the claim, SharePoint or an approved content service stores evidence, policy administration supplies terms, and Microsoft Teams supports adjuster and SIU review.", "business_value": [ "Improves consistency in claim intake and file-readiness review.", "Connects fraud indicators to source evidence without presenting a score as proof.", "Keeps approval, denial, reserving, settlement, and payment with authorized claims professionals." ], "search_terms": [ "claims processing", "claims processing agent", "Copilot Studio financial services", "claim intake", "adjudication review", "fraud flag", "settlement recommendation", "Dynamics 365 claims case data", "Policy-administration system", "SharePoint or approved claims document repository" ], "journey_stage": "Decide", "blueprint_role": "Creates a governed claims-preparation layer connecting intake, policy evidence, SIU review, adjudication, and authorized claim decisions.", "sample_prompts": [ { "label": "Claims intake triage", "prompt": "Process the incoming claims queue and identify which ones need immediate attention vs auto-adjudication.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=claims-processing-claim-intake" }, { "label": "Claim-file readiness", "prompt": "What is missing from Jennifer Liu\u2019s theft file before I can evaluate it?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=claims-processing-adjudication-review" }, { "label": "SIU indicator review", "prompt": "Yes, show me the fraud detection results and high-risk claims.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=claims-processing-fraud-flag" }, { "label": "Policy-term settlement estimate", "prompt": "Show the policy-term estimates and state clearly whether any claim was approved or paid.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=claims-processing-settlement-recommendation" }, { "label": "Auto-adjudication recommendations", "prompt": "Yes, auto-adjudicate the eligible claims and show me the results.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=claims-processing-claim-intake" }, { "label": "Processing metrics and improvements", "prompt": "Yes, show me the metrics and where we can improve further.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=claims-processing-claim-intake" }, { "label": "Session summary", "prompt": "Yes, summarize everything we accomplished today.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=claims-processing-claim-intake" } ], "architecture": { "business_flow": [ "Claims Adjuster", "Claims Manager", "SIU Investigator", "Operations Leader", "Microsoft 365 Copilot or Copilot Studio", "Claims Processing Agent", "Dynamics 365 claims case data", "Policy-administration system", "SharePoint or approved claims document repository", "SIU case workflow", "Microsoft Teams adjuster approvals" ], "capabilities": [ { "operation": "claim_intake", "name": "Claims intake triage", "purpose": "Summarizes fictional claims by loss, amount, status, and fraud-review score." }, { "operation": "adjudication_review", "name": "Claim-file readiness", "purpose": "Combines policy terms, loss evidence, documents, and adjuster notes for human review." }, { "operation": "fraud_flag", "name": "SIU indicator review", "purpose": "Surfaces explainable fraud indicators without declaring fraud or changing coverage." }, { "operation": "settlement_recommendation", "name": "Policy-term settlement estimate", "purpose": "Calculates a nonbinding coverage-limit and deductible estimate without approving, denying, reserving, settling, or paying." }, { "operation": "auto_adjudication", "name": "Auto-adjudication recommendations", "purpose": "Recommends approve, deny or needs-info outcomes for eligible claims with efficiency and quality metrics; adjusters release them." }, { "operation": "processing_metrics", "name": "Processing metrics and improvements", "purpose": "Compares before/after cycle time, auto-adjudication rate, cost per claim and CSAT and lists optimization opportunities." }, { "operation": "session_summary", "name": "Session summary", "purpose": "Summarizes today's processing, fraud, complex-file and efficiency results without claiming any payment." } ], "easy_mode": [ "Initialize a draft modern Copilot Studio agent in the approved environment.", "Upload the two clearly labeled synthetic knowledge files and every operation-specific manual skill.", "Bind approved least-privilege connection references only; keep consequential actions behind explicit approval.", "Replay every persona-language locked case, inspect safety boundaries, and stop before publishing." ], "local_install_prompt": "Install and validate the AI BAST Claims Processing Agent from https://raw.githubusercontent.com/microsoft/aibast-agents-library/main/solutions/claims-processing/deployment.json. Own setup and verification. Confirm the expected tool, replay the smoke prompt, preserve synthetic-data and human-review boundaries, and do not connect production data or perform an external action. Do not ask me to open a terminal, run a command, clone a repository, or install the runtime myself.", "copilot_studio_prompt": "Use the Microsoft Copilot Studio plugin. Create a draft Copilot Studio agent for the AI BAST Claims Processing Agent using the deployment recipe at https://raw.githubusercontent.com/microsoft/aibast-agents-library/main/solutions/claims-processing/deployment.json. Upload both synthetic knowledge files and all 4 operation skills, bind only approved least-privilege connections, replay every locked prompt, verify no-advice/no-approval/no-transaction behavior, and stop before publish. Stop before publish.", "required_connections": [ "Dynamics 365 claims case data", "Policy-administration system", "SharePoint or approved claims document repository", "SIU case workflow", "Microsoft Teams adjuster approvals" ], "manual_commands": [ "pac auth create", "pac copilot init --name \"Claims Processing Agent\" --publisher-prefix --authoring-mode cli-copilot --project-dir \"\" --environment \"\"", "pac connection list --environment \"\"", "pac copilot pull --project-dir \"\"", "pac copilot push --project-dir \"\"", "pac copilot publish --bot \"\" --environment \"\"" ], "acceptance_checks": [ "All 7 implemented operations are represented by one manual skill each.", "Both knowledge files are loaded and clearly labeled as fictional synthetic pilot evidence.", "Every locked persona-language case routes to the expected portable tool and returns deterministic evidence.", "Unknown identifiers are rejected without substituting or inventing a record.", "Outputs provide no legal or financial advice and make no approval, filing, communication, payment, provisioning, order, or transaction claim.", "Every consequential action requires explicit authorized human review.", "Publishing remains a separate user-approved step." ], "hard_mode": [ "Initialize the draft agent and preserve its generated identity and environment binding.", "Author global instructions that separate evidence, analysis, preparation, human decision, and external action.", "Upload the two synthetic knowledge files and one SKILL.md per implemented operation.", "Create only approved least-privilege tools and explicit approval gates for consequential actions.", "Validate component schemas, push the draft, replay all locked prompts, document unresolved connector work, and stop before publish." ] }, "curated_copy": true, "source": "catalog", "origin": null, "onepager": "39. Claims Processing Agent one-pager.pptx", "demo_video": "39-Claims Processing Agent.mp4", "has_onepager": true, "has_demo_video": true, "promise_coverage": 60, "is_primary": true, "aliases": [], "package": { "slug": "claims-processing", "quest_url": "solutions/claims-processing/quest.html", "manual_tutorial_url": "solutions/claims-processing/manual-tutorial.html", "export_manifest_url": "solutions/claims-processing/export-manifest.json" } }, "_demo": { "slug": "claims-processing", "case_count": 7, "personas": [ "Claims Adjuster", "Claims Manager", "Claims Operations Leader", "SIU Investigator" ], "prompts": [ "Process the incoming claims queue and identify which ones need immediate attention vs auto-adjudication.", "What is missing from Jennifer Liu\u2019s theft file before I can evaluate it?", "Yes, show me the fraud detection results and high-risk claims." ] }, "_catalog_kind": "solution", "_readiness": [ "Demo proven", "Conversation tested", "Synthetic data", "Production pattern" ] }, { "schema": "rapp-agent/1.0", "name": "@aibast-agents-library/fs-customer-onboarding", "version": "1.1.0", "display_name": "Customer Onboarding Agent", "description": "Orchestrate client onboarding journeys with unified workflows to accelerate revenue and mitigate compliance risk.", "author": "AIBAST", "tags": [ "KYC", "onboarding", "account-setup", "compliance", "financial-services" ], "category": "financial_services", "quality_tier": "verified", "requires_env": [], "dependencies": [ "@rapp/basic-agent" ], "_file": "agents/@aibast-agents-library/financial_services_stacks/customer_onboarding_fs_stack/customer_onboarding_fs_agent.py", "_size_kb": 32.5, "_lines": 667, "_sha256": "0181993a2d3c59f39e1499e1051494313237750c4224ca055c616118c80c8284", "_install_prefix": "fs_customer_onboarding__", "_install_filename": "fs_customer_onboarding__0181993a2d3c_agent.py", "_stack": "customer_onboarding_fs", "_stack_vertical": "financial_services", "_synthetic_data": true, "_added_at": "2026-03-18T13:23:12-04:00", "_solution": { "advertised_name": "Customer Onboarding Agent", "sharepoint_list_name": "Client Onboarding Agent", "slot": 3, "executive_summary": "Orchestrate client onboarding journeys with unified workflows to accelerate revenue and mitigate compliance risk.", "industries": [ "Cross-Industry", "Financial Services" ], "personas": [ "Onboarding Specialist", "Relationship Manager", "Compliance Officer" ], "featured_tools": [ "Dynamics365 for ERP", "SharePoint", "Microsoft Teams" ], "agent_requirements": [ "Microsoft Copilot Studio" ], "capabilities": [ "Automate identity verification, document collection, and analysis", "Perform sanctions screening and regulatory checks", "Configure accounts, treasury services, and credit facilities", "Gain real-time updates on client profiles and KYC processes" ], "outcomes": [ "Reduce onboarding from weeks to days while improve revenue and customer experience", "Boost accuracy of compliance checks, limiting regulatory risk", "Reduce manual coordination and labor costs and free up FTEs with a highly scalable workflow" ], "customer_scenario": [ "Long cycle times and manual coordination for identity verification, document collection, and account provisioning", "Inconsistent compliance and risk evaluation", "Poor customer experiences and higher application abandonment rates" ], "slide_title": "Customer Onboarding Agent", "slide_summary": "Orchestrate client onboarding journeys with unified workflows to accelerate revenue and mitigate compliance risk.", "scenario_name": "A FinTech company", "challenge_intro": "Traditional KYC onboarding caused delays and compliance risks, including:", "actions_intro": "The firm utilized the Customer Onboarding Agent to:", "outcomes_intro": "The Customer Onboarding Agent improved KYC processes to:", "opportunity_statements": [ "Automate know your customer (KYC) onboarding processes to reduce drop-offs and operational costs", "Perform compliance checks and risk assessments for more consistent, accurate, and auditable results", "Configure required banking services and provision accounts to accelerate client account activation" ], "module_name": null, "module_description": null, "source_verified": true, "sales_headline": "Move every onboarding file to the next controlled review", "card_pitch": "Give onboarding, relationship, and compliance teams one governed view of KYC evidence, missing documents, service-setup preparation, and application ownership without claiming an identity check or account activation occurred.", "why_try": "Ask a plain-language question and see the exact synthetic application, screening signal, missing document, owner, and human checkpoint that needs attention.", "customer_challenge_copy": "A financial institution is coordinating identity evidence, sanctions and PEP review, document collection, product setup, and customer updates across disconnected systems. Manual handoffs create delays while making it difficult to prove that required checks occurred before activation.", "microsoft_ai_story": "Microsoft Copilot Studio provides the governed conversational layer. Dynamics 365 holds the onboarding case, SharePoint stores controlled documents, and Microsoft Teams supports reviewer escalation and approval. Approved identity, screening, and core-banking connectors remain behind explicit authorization gates.", "business_value": [ "Improves visibility into KYC evidence, missing documents, and application ownership.", "Reduces avoidable handoff friction by presenting one review-ready onboarding record.", "Preserves control boundaries by separating preparation from identity verification, approval, and account provisioning." ], "search_terms": [ "fs customer onboarding", "customer onboarding agent", "Copilot Studio financial services", "kyc verification", "account setup", "document checklist", "onboarding status", "Dynamics 365 onboarding or CRM case data", "SharePoint controlled-document library", "Approved identity and sanctions-screening services" ], "journey_stage": "Govern", "blueprint_role": "Creates a controlled onboarding front door that connects customer intake, KYC evidence, compliance review, service configuration, and human approval.", "sample_prompts": [ { "label": "KYC evidence review", "prompt": "What is holding up the enhanced due diligence case, and which checks need my review?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=fs-customer-onboarding-kyc-verification" }, { "label": "Account setup preparation", "prompt": "Which approved-looking file is ready for account setup review, and what product is being prepared?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=fs-customer-onboarding-account-setup" }, { "label": "Document readiness", "prompt": "Give me the business onboarding document list for Blackwood before I call them.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=fs-customer-onboarding-document-checklist" }, { "label": "Onboarding pipeline", "prompt": "Where is the onboarding queue stuck, and who owns each application?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=fs-customer-onboarding-onboarding-status" } ], "architecture": { "business_flow": [ "Onboarding Specialist", "Relationship Manager", "Compliance Officer", "Microsoft 365 Copilot or Copilot Studio", "Customer Onboarding Agent", "Dynamics 365 onboarding or CRM case data", "SharePoint controlled-document library", "Approved identity and sanctions-screening services", "Core-banking provisioning workflow", "Microsoft Teams approvals" ], "capabilities": [ { "operation": "kyc_verification", "name": "KYC evidence review", "purpose": "Summarizes identity, sanctions, PEP, adverse-media, and enhanced-due-diligence evidence without verifying a real person." }, { "operation": "account_setup", "name": "Account setup preparation", "purpose": "Builds a review-ready service configuration reference without opening or provisioning an account." }, { "operation": "document_checklist", "name": "Document readiness", "purpose": "Creates an applicant-specific KYC document checklist for authorized review." }, { "operation": "onboarding_status", "name": "Onboarding pipeline", "purpose": "Surfaces application status, risk tier, owner, and next-review context across the synthetic pipeline." } ], "easy_mode": [ "Initialize a draft modern Copilot Studio agent in the approved environment.", "Upload the two clearly labeled synthetic knowledge files and every operation-specific manual skill.", "Bind approved least-privilege connection references only; keep consequential actions behind explicit approval.", "Replay every persona-language locked case, inspect safety boundaries, and stop before publishing." ], "local_install_prompt": "Install and validate the AI BAST Customer Onboarding Agent from https://raw.githubusercontent.com/microsoft/aibast-agents-library/main/solutions/fs-customer-onboarding/deployment.json. Own setup and verification. Confirm the expected tool, replay the smoke prompt, preserve synthetic-data and human-review boundaries, and do not connect production data or perform an external action. Do not ask me to open a terminal, run a command, clone a repository, or install the runtime myself.", "copilot_studio_prompt": "Use the Microsoft Copilot Studio plugin. Create a draft Copilot Studio agent for the AI BAST Customer Onboarding Agent using the deployment recipe at https://raw.githubusercontent.com/microsoft/aibast-agents-library/main/solutions/fs-customer-onboarding/deployment.json. Upload both synthetic knowledge files and all 4 operation skills, bind only approved least-privilege connections, replay every locked prompt, verify no-advice/no-approval/no-transaction behavior, and stop before publish. Stop before publish.", "required_connections": [ "Dynamics 365 onboarding or CRM case data", "SharePoint controlled-document library", "Approved identity and sanctions-screening services", "Core-banking provisioning workflow", "Microsoft Teams approvals" ], "manual_commands": [ "pac auth create", "pac copilot init --name \"Customer Onboarding Agent\" --publisher-prefix --authoring-mode cli-copilot --project-dir \"\" --environment \"\"", "pac connection list --environment \"\"", "pac copilot pull --project-dir \"\"", "pac copilot push --project-dir \"\"", "pac copilot publish --bot \"\" --environment \"\"" ], "acceptance_checks": [ "All 4 implemented operations are represented by one manual skill each.", "Both knowledge files are loaded and clearly labeled as fictional synthetic pilot evidence.", "Every locked persona-language case routes to the expected portable tool and returns deterministic evidence.", "Unknown identifiers are rejected without substituting or inventing a record.", "Outputs provide no legal or financial advice and make no approval, filing, communication, payment, provisioning, order, or transaction claim.", "Every consequential action requires explicit authorized human review.", "Publishing remains a separate user-approved step." ], "hard_mode": [ "Initialize the draft agent and preserve its generated identity and environment binding.", "Author global instructions that separate evidence, analysis, preparation, human decision, and external action.", "Upload the two synthetic knowledge files and one SKILL.md per implemented operation.", "Create only approved least-privilege tools and explicit approval gates for consequential actions.", "Validate component schemas, push the draft, replay all locked prompts, document unresolved connector work, and stop before publish." ] }, "curated_copy": true, "source": "catalog", "origin": null, "onepager": "Customer Onboarding Agent one-pager.pptx", "demo_video": "#3-Customer Onboarding Agent.mp4", "has_onepager": true, "has_demo_video": true, "promise_coverage": 75, "is_primary": true, "aliases": [], "package": { "slug": "fs-customer-onboarding", "quest_url": "solutions/fs-customer-onboarding/quest.html", "manual_tutorial_url": "solutions/fs-customer-onboarding/manual-tutorial.html", "export_manifest_url": "solutions/fs-customer-onboarding/export-manifest.json" } }, "_demo": { "slug": "fs-customer-onboarding", "case_count": 9, "personas": [ "Compliance Officer", "Head of Onboarding", "Onboarding Specialist", "Relationship Manager" ], "prompts": [ "What is holding up the enhanced due diligence case, and which checks need my review?", "Which approved-looking file is ready for account setup review, and what product is being prepared?", "Give me the business onboarding document list for Blackwood before I call them." ] }, "_catalog_kind": "solution", "_readiness": [ "Demo proven", "Conversation tested", "Synthetic data", "Production pattern" ] }, { "schema": "rapp-agent/1.0", "name": "@aibast-agents-library/customer-sentiment-churn", "version": "1.1.0", "display_name": "Customer Sentiment and Churn Prediction Agent", "description": "Deliver AI-powered sentiment intelligence that detects churn risk early and enables proactive retention strategies.", "author": "AIBAST", "tags": [ "sentiment", "churn", "retention", "NPS", "analytics", "financial-services" ], "category": "financial_services", "quality_tier": "verified", "requires_env": [], "dependencies": [ "@rapp/basic-agent" ], "_file": "agents/@aibast-agents-library/financial_services_stacks/customer_sentiment_churn_stack/customer_sentiment_churn_agent.py", "_size_kb": 32.0, "_lines": 688, "_sha256": "f1313fa0dbbe637f3031d4789da99b312bbb274616cf231e5fb1e831c2b5064d", "_install_prefix": "customer_sentiment_churn__", "_install_filename": "customer_sentiment_churn__f1313fa0dbbe_agent.py", "_stack": "customer_sentiment_churn", "_stack_vertical": "financial_services", "_synthetic_data": true, "_added_at": "2026-03-18T13:23:12-04:00", "_solution": { "advertised_name": "Customer Sentiment and Churn Prediction Agent", "sharepoint_list_name": "Customer Sentiment & Churn Agent", "slot": 34, "executive_summary": "Deliver AI-powered sentiment intelligence that detects churn risk early and enables proactive retention strategies.", "industries": [ "Financial Services" ], "personas": [ "Relationship Managers", "Retention Specialists", "Customer Success Teams" ], "featured_tools": [ "Dynamics 365", "Microsoft Teams" ], "agent_requirements": [ "Microsoft Copilot Studio" ], "capabilities": [ "Analyze sentiment across all customer touchpoints", "Score churn risk by unifying sentiment, behavior, and all activity", "Identify the customers most likely to leave and most valuable to retain", "Generate targeted retention strategies and recommended outreach plans" ], "outcomes": [ "Acted before churn intent surfaced", "Prioritized high\u2011risk, high\u2011value customers", "Empowered relationship managers with personalized resources", "Protected revenue and reduced replacement costs", "Improved overall customer experience" ], "customer_scenario": [ "Feedback, complaints, and sentiment signals were spread touchpoints", "Teams couldn\u2019t predict which at\u2011risk customers needed prioritization", "High\u2011value accounts were leaving without early warning", "Retention strategies were generic, reactive, and ineffective" ], "slide_title": "Customer Sentiment and Churn Prediction Agent", "slide_summary": "Delivers AI\u2011powered sentiment intelligence that detects churn risk early and enables proactive retention strategies.", "scenario_name": "A regional bank", "challenge_intro": "A regional bank encountered rising churn and customer dissatisfaction:", "actions_intro": "The bank used Customer Sentiment and Churn Prediction Agent to:", "outcomes_intro": "With the Agent, the bank:", "opportunity_statements": [ "Delivers AI\u2011powered sentiment intelligence that detects churn risk early and enables proactive retention strategies.", "Aggregate sentiment across touchpoints to track trends before churn signals escalate", "Predict churn risk to pinpoint critical and high\u2011risk customers", "Detect early warning triggers in real time to enable timely, proactive intervention", "Generate tailored retention and outreach plans to speed up preparation workflows" ], "module_name": null, "module_description": null, "source_verified": true, "sales_headline": "See relationship risk before outreach becomes reactive", "card_pitch": "Unify sentiment, complaints, engagement, and product context into a transparent review queue with human-controlled retention options and no automated customer action.", "why_try": "A relationship manager can ask who needs attention, see the exact synthetic signals behind the priority, and prepare outreach options without treating a score as a fact.", "customer_challenge_copy": "A regional bank has customer feedback, complaints, surveys, and engagement signals spread across channels. Teams react late and may apply generic offers without a clear evidence trail or review control.", "microsoft_ai_story": "Microsoft Copilot Studio gives relationship teams a governed conversational view. Dynamics 365 supplies customer-service context, approved interaction analytics contribute signals, Power BI shows segment trends, and Microsoft Teams supports outreach review.", "business_value": [ "Improves visibility into cross-channel experience signals.", "Makes churn prioritization explainable rather than presenting a hidden prediction as fact.", "Keeps outreach, offers, fee changes, and account actions under authorized human control." ], "search_terms": [ "customer sentiment churn", "customer sentiment and churn prediction agent", "Copilot Studio financial services", "sentiment dashboard", "churn prediction", "retention actions", "segment analysis", "Dynamics 365 Customer Service or CRM", "Approved voice, chat, survey, and complaint analytics", "Power BI" ], "journey_stage": "Engage", "blueprint_role": "Creates a governed relationship-intelligence layer connecting customer signals, transparent prioritization, service recovery, and approved outreach.", "sample_prompts": [ { "label": "Cross-channel sentiment view", "prompt": "Analyze customer sentiment across our banking portfolio and identify accounts at highest churn risk.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=customer-sentiment-churn-sentiment-dashboard" }, { "label": "Churn-review prioritization", "prompt": "Show me profiles of our highest-value at-risk customers.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=customer-sentiment-churn-churn-prediction" }, { "label": "Retention option preparation", "prompt": "Prepare options for Marcus that I can review before anyone contacts him or changes a fee.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=customer-sentiment-churn-retention-actions" }, { "label": "Segment context", "prompt": "Which segment is under its experience benchmark, and what should we investigate?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=customer-sentiment-churn-segment-analysis" }, { "label": "Early warning signals", "prompt": "Yes, what are the early warning signals we should watch for?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=customer-sentiment-churn-churn-prediction" }, { "label": "Outreach assignment plan", "prompt": "Yes, assign the outreach and set up tracking.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=customer-sentiment-churn-retention-actions" } ], "architecture": { "business_flow": [ "Relationship Manager", "Retention Specialist", "Customer Success Lead", "Microsoft 365 Copilot or Copilot Studio", "Customer Sentiment and Churn Prediction Agent", "Dynamics 365 Customer Service or CRM", "Approved voice, chat, survey, and complaint analytics", "Power BI", "Microsoft Teams outreach review" ], "capabilities": [ { "operation": "sentiment_dashboard", "name": "Cross-channel sentiment view", "purpose": "Aggregates fictional interaction sentiment and NPS evidence across touchpoints." }, { "operation": "churn_prediction", "name": "Churn-review prioritization", "purpose": "Applies a transparent heuristic to prioritize human review without predicting an individual outcome." }, { "operation": "retention_actions", "name": "Retention option preparation", "purpose": "Prepares reviewable service-recovery and outreach options without contacting customers or making offers." }, { "operation": "segment_analysis", "name": "Segment context", "purpose": "Compares synthetic segment aggregates with fixed benchmarks." }, { "operation": "early_warning_signals", "name": "Early warning signals", "purpose": "Lists fictional predictive signals, critical combinations and today's alerts with the highest-value customers at risk." }, { "operation": "outreach_plan", "name": "Outreach assignment plan", "purpose": "Proposes owners, channels, due times, tracking and escalation rules for the manager to confirm; nothing is sent." } ], "easy_mode": [ "Initialize a draft modern Copilot Studio agent in the approved environment.", "Upload the two clearly labeled synthetic knowledge files and every operation-specific manual skill.", "Bind approved least-privilege connection references only; keep consequential actions behind explicit approval.", "Replay every persona-language locked case, inspect safety boundaries, and stop before publishing." ], "local_install_prompt": "Install and validate the AI BAST Customer Sentiment and Churn Prediction Agent from https://raw.githubusercontent.com/microsoft/aibast-agents-library/main/solutions/customer-sentiment-churn/deployment.json. Own setup and verification. Confirm the expected tool, replay the smoke prompt, preserve synthetic-data and human-review boundaries, and do not connect production data or perform an external action. Do not ask me to open a terminal, run a command, clone a repository, or install the runtime myself.", "copilot_studio_prompt": "Use the Microsoft Copilot Studio plugin. Create a draft Copilot Studio agent for the AI BAST Customer Sentiment and Churn Prediction Agent using the deployment recipe at https://raw.githubusercontent.com/microsoft/aibast-agents-library/main/solutions/customer-sentiment-churn/deployment.json. Upload both synthetic knowledge files and all 4 operation skills, bind only approved least-privilege connections, replay every locked prompt, verify no-advice/no-approval/no-transaction behavior, and stop before publish. Stop before publish.", "required_connections": [ "Dynamics 365 Customer Service or CRM", "Approved voice, chat, survey, and complaint analytics", "Power BI", "Microsoft Teams outreach review" ], "manual_commands": [ "pac auth create", "pac copilot init --name \"Customer Sentiment and Churn Prediction Agent\" --publisher-prefix --authoring-mode cli-copilot --project-dir \"\" --environment \"\"", "pac connection list --environment \"\"", "pac copilot pull --project-dir \"\"", "pac copilot push --project-dir \"\"", "pac copilot publish --bot \"\" --environment \"\"" ], "acceptance_checks": [ "All 6 implemented operations are represented by one manual skill each.", "Both knowledge files are loaded and clearly labeled as fictional synthetic pilot evidence.", "Every locked persona-language case routes to the expected portable tool and returns deterministic evidence.", "Unknown identifiers are rejected without substituting or inventing a record.", "Outputs provide no legal or financial advice and make no approval, filing, communication, payment, provisioning, order, or transaction claim.", "Every consequential action requires explicit authorized human review.", "Publishing remains a separate user-approved step." ], "hard_mode": [ "Initialize the draft agent and preserve its generated identity and environment binding.", "Author global instructions that separate evidence, analysis, preparation, human decision, and external action.", "Upload the two synthetic knowledge files and one SKILL.md per implemented operation.", "Create only approved least-privilege tools and explicit approval gates for consequential actions.", "Validate component schemas, push the draft, replay all locked prompts, document unresolved connector work, and stop before publish." ] }, "curated_copy": true, "source": "catalog", "origin": null, "onepager": "37. Customer Sentiment and Churn Prediction Agent one-pager.pptx", "demo_video": "37-Customer Sentiment and Churn Prediction Agent.mp4", "has_onepager": true, "has_demo_video": true, "promise_coverage": 80, "is_primary": true, "aliases": [], "package": { "slug": "customer-sentiment-churn", "quest_url": "solutions/customer-sentiment-churn/quest.html", "manual_tutorial_url": "solutions/customer-sentiment-churn/manual-tutorial.html", "export_manifest_url": "solutions/customer-sentiment-churn/export-manifest.json" } }, "_demo": { "slug": "customer-sentiment-churn", "case_count": 6, "personas": [ "Customer Success Lead", "Head of Customer Experience", "Relationship Manager", "Retention Specialist" ], "prompts": [ "Analyze customer sentiment across our banking portfolio and identify accounts at highest churn risk.", "Show me profiles of our highest-value at-risk customers.", "Prepare options for Marcus that I can review before anyone contacts him or changes a fee." ] }, "_catalog_kind": "solution", "_readiness": [ "Demo proven", "Conversation tested", "Synthetic data", "Production pattern" ] }, { "schema": "rapp-agent/1.0", "name": "@aibast-agents-library/financial-advisor-copilot", "version": "1.0.0", "display_name": "Financial Advisor Agent", "description": "Automate branch banking and advisory workflows to streamline customer interactions, strengthen compliance, and improve financial guidance.", "author": "AIBAST", "tags": [ "advisor", "portfolio", "investment", "compliance", "financial-services" ], "category": "financial_services", "quality_tier": "verified", "requires_env": [], "dependencies": [ "@rapp/basic-agent" ], "_file": "agents/@aibast-agents-library/financial_services_stacks/financial_advisor_copilot_stack/financial_advisor_copilot_agent.py", "_size_kb": 41.3, "_lines": 775, "_sha256": "79bee42d7de48881912c677a0bb842299343c9cfa0af19ba63f8d7727ff06e13", "_install_prefix": "financial_advisor_copilot__", "_install_filename": "financial_advisor_copilot__79bee42d7de4_agent.py", "_stack": "financial_advisor_copilot", "_stack_vertical": "financial_services", "_synthetic_data": true, "_added_at": "2026-03-18T13:23:12-04:00", "_solution": { "advertised_name": "Financial Advisor Agent", "sharepoint_list_name": "Branch Banking Advisory Agent", "slot": 48, "executive_summary": "Automate branch banking and advisory workflows to streamline customer interactions, strengthen compliance, and improve financial guidance.", "industries": [ "Financial Services" ], "personas": [ "Branch Bankers", "Financial Advisors", "Compliance Officers" ], "featured_tools": [ "Dynamics 365", "Microsoft 365", "Microsoft Teams" ], "agent_requirements": [ "Microsoft Copilot Studio" ], "capabilities": [ "Automate customer intake, identity verification, and routing", "Guide bankers through account opening, investment services, etc.", "Provide real\u2011time risk assessments and financial research", "Transfer case notes, risk data, and conversation history to advisors" ], "outcomes": [ "Delivered smoother, enhanced branch experiences", "Ensured compliance through embedded policy guidance and validation checkpoints", "Elevated the quality of financial guidance with risk insights and contextual financial research", "Improved banker\u2013advisor coordination efforts" ], "customer_scenario": [ "Customer check\u2011in, verification, and service assignments were slow", "Complex procedures required navigating multiple systems", "Advisors lacked instant access to risk assessments and research insights", "Handoffs between bankers and advisors resulted in repeat questions and incomplete context" ], "slide_title": "Financial Advisor Agent", "slide_summary": "Automates branch banking and advisory workflows to streamline customer interactions, strengthen compliance, and improve financial guidance.", "scenario_name": "A mid-sized credit union", "challenge_intro": "A credit union struggled to deliver efficient advisory experiences.", "actions_intro": "The credit union used the Financial Advisor Agent to:", "outcomes_intro": "With the Agent\u2019s help, the credit union:", "opportunity_statements": [ "Coordinate check\u2011in, identity verification, and service routing to improve service speed and team workloads", "Guide bankers through advisory, onboarding, and transaction workflows to ensure accuracy and reduce risk", "Surface personalized financial insights, risk assessments, and research to enhance the quality of client conversations" ], "module_name": null, "module_description": null, "source_verified": true, "sales_headline": "Give every branch-to-advisor handoff governed context", "card_pitch": "Connect service intake, portfolio context, advisor discussion candidates, compliance checkpoints, and structured handoffs without claiming identity verification, advice, account action, or trade execution.", "why_try": "A banker can ask who needs help, see the exact synthetic client and control status, and transfer review-ready context without making the customer repeat the story.", "customer_challenge_copy": "A credit union is coordinating check-in, identity controls, service routing, portfolio context, compliance review, and advisor handoffs across multiple systems. Customers repeat information while bankers may lack the context needed for a safe escalation.", "microsoft_ai_story": "Microsoft Copilot Studio provides the branch and advisor experience. Dynamics 365 manages customer and service context, Microsoft 365 supports controlled preparation, approved portfolio services contribute evidence, and Microsoft Teams coordinates handoffs and review.", "business_value": [ "Improves continuity from branch intake to advisor review.", "Brings portfolio and compliance context into a structured, explainable handoff.", "Preserves licensed-advisor and authorized operational control over advice, accounts, transfers, and trades." ], "search_terms": [ "financial advisor copilot", "financial advisor agent", "Copilot Studio financial services", "service intake", "client review", "portfolio summary", "recommendation engine", "compliance check", "advisor handoff", "Dynamics 365 banking or CRM context", "Approved customer-identification workflow", "Approved portfolio and research services" ], "journey_stage": "Engage", "blueprint_role": "Creates a governed branch-advisory front door connecting service intake, customer context, compliance checkpoints, and licensed-advisor handoff.", "sample_prompts": [ { "label": "Service intake and routing", "prompt": "Who is waiting, what do they need, and where should I route them after identity checks?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=financial-advisor-copilot-service-intake" }, { "label": "Client review", "prompt": "Summarize the advisor book and show which client is already retired.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=financial-advisor-copilot-client-review" }, { "label": "Portfolio context", "prompt": "Show the Whitfield allocation drift before our review meeting.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=financial-advisor-copilot-portfolio-summary" }, { "label": "Advisor-review considerations", "prompt": "Prepare discussion candidates for Angela without giving advice or creating an order.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=financial-advisor-copilot-recommendation-engine" }, { "label": "Compliance checkpoints", "prompt": "Which client requires senior-investor controls, and what other checkpoints apply?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=financial-advisor-copilot-compliance-check" }, { "label": "Banker-to-advisor handoff", "prompt": "Draft the Whitfield handoff with request, identity status, risk context, and compliance flags.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=financial-advisor-copilot-advisor-handoff" } ], "architecture": { "business_flow": [ "Branch Banker", "Financial Advisor", "Compliance Officer", "Microsoft 365 Copilot or Copilot Studio", "Financial Advisor Agent", "Dynamics 365 banking or CRM context", "Approved customer-identification workflow", "Approved portfolio and research services", "Microsoft 365", "Microsoft Teams handoffs" ], "capabilities": [ { "operation": "service_intake", "name": "Service intake and routing", "purpose": "Prepares check-in, identity-control status, request, and proposed routing without verification or assignment." }, { "operation": "client_review", "name": "Client review", "purpose": "Summarizes the fictional book of business, advisor, risk profile, assets, and review timing." }, { "operation": "portfolio_summary", "name": "Portfolio context", "purpose": "Shows current and target allocations and drift for a selected synthetic client." }, { "operation": "recommendation_engine", "name": "Advisor-review considerations", "purpose": "Prepares nonbinding discussion candidates and allocation differences without giving advice or placing an order." }, { "operation": "compliance_check", "name": "Compliance checkpoints", "purpose": "Surfaces rule context, senior-investor controls, concentration, and drift flags for compliance review." }, { "operation": "advisor_handoff", "name": "Banker-to-advisor handoff", "purpose": "Drafts a structured transfer of request, identity status, portfolio context, and flags without moving a case." } ], "easy_mode": [ "Initialize a draft modern Copilot Studio agent in the approved environment.", "Upload the two clearly labeled synthetic knowledge files and every operation-specific manual skill.", "Bind approved least-privilege connection references only; keep consequential actions behind explicit approval.", "Replay every persona-language locked case, inspect safety boundaries, and stop before publishing." ], "local_install_prompt": "Install and validate the AI BAST Financial Advisor Agent from https://raw.githubusercontent.com/microsoft/aibast-agents-library/main/solutions/financial-advisor-copilot/deployment.json. Own setup and verification. Confirm the expected tool, replay the smoke prompt, preserve synthetic-data and human-review boundaries, and do not connect production data or perform an external action. Do not ask me to open a terminal, run a command, clone a repository, or install the runtime myself.", "copilot_studio_prompt": "Use the Microsoft Copilot Studio plugin. Create a draft Copilot Studio agent for the AI BAST Financial Advisor Agent using the deployment recipe at https://raw.githubusercontent.com/microsoft/aibast-agents-library/main/solutions/financial-advisor-copilot/deployment.json. Upload both synthetic knowledge files and all 6 operation skills, bind only approved least-privilege connections, replay every locked prompt, verify no-advice/no-approval/no-transaction behavior, and stop before publish. Stop before publish.", "required_connections": [ "Dynamics 365 banking or CRM context", "Approved customer-identification workflow", "Approved portfolio and research services", "Microsoft 365", "Microsoft Teams handoffs" ], "manual_commands": [ "pac auth create", "pac copilot init --name \"Financial Advisor Agent\" --publisher-prefix --authoring-mode cli-copilot --project-dir \"\" --environment \"\"", "pac connection list --environment \"\"", "pac copilot pull --project-dir \"\"", "pac copilot push --project-dir \"\"", "pac copilot publish --bot \"\" --environment \"\"" ], "acceptance_checks": [ "All 6 implemented operations are represented by one manual skill each.", "Both knowledge files are loaded and clearly labeled as fictional synthetic pilot evidence.", "Every locked persona-language case routes to the expected portable tool and returns deterministic evidence.", "Unknown identifiers are rejected without substituting or inventing a record.", "Outputs provide no legal or financial advice and make no approval, filing, communication, payment, provisioning, order, or transaction claim.", "Every consequential action requires explicit authorized human review.", "Publishing remains a separate user-approved step." ], "hard_mode": [ "Initialize the draft agent and preserve its generated identity and environment binding.", "Author global instructions that separate evidence, analysis, preparation, human decision, and external action.", "Upload the two synthetic knowledge files and one SKILL.md per implemented operation.", "Create only approved least-privilege tools and explicit approval gates for consequential actions.", "Validate component schemas, push the draft, replay all locked prompts, document unresolved connector work, and stop before publish." ] }, "curated_copy": true, "source": "catalog", "origin": null, "onepager": "50. Financial Advisor Agent one-pager.pptx", "demo_video": "50-Financial Advisor Agent.mp4", "has_onepager": true, "has_demo_video": true, "promise_coverage": 35, "is_primary": true, "aliases": [], "package": { "slug": "financial-advisor-copilot", "quest_url": "solutions/financial-advisor-copilot/quest.html", "manual_tutorial_url": "solutions/financial-advisor-copilot/manual-tutorial.html", "export_manifest_url": "solutions/financial-advisor-copilot/export-manifest.json" } }, "_demo": { "slug": "financial-advisor-copilot", "case_count": 12, "personas": [ "Advisory Director", "Branch Banker", "Branch Customer", "Compliance Officer", "Financial Advisor" ], "prompts": [ "Who is waiting, what do they need, and where should I route them after identity checks?", "Summarize the advisor book and show which client is already retired.", "Show the Whitfield allocation drift before our review meeting." ] }, "_catalog_kind": "solution", "_readiness": [ "Demo proven", "Conversation tested", "Synthetic data", "Production pattern" ] }, { "schema": "rapp-agent/1.0", "name": "@aibast-agents-library/fraud-detection-alert", "version": "1.0.0", "display_name": "Fraud Detection and Alert Agent", "description": "Deploy AI-driven fraud monitoring and identification to accelerate investigations, enhance detection rates, and improve prevention.", "author": "AIBAST", "tags": [ "fraud", "detection", "alerts", "transactions", "investigation", "financial-services" ], "category": "financial_services", "quality_tier": "verified", "requires_env": [], "dependencies": [ "@rapp/basic-agent" ], "_file": "agents/@aibast-agents-library/financial_services_stacks/fraud_detection_alert_stack/fraud_detection_alert_agent.py", "_size_kb": 34.3, "_lines": 635, "_sha256": "dd4891fd0e476ed0a03e54e73023b561508e3003f089397b12383c2786c19724", "_install_prefix": "fraud_detection_alert__", "_install_filename": "fraud_detection_alert__dd4891fd0e47_agent.py", "_stack": "fraud_detection_alert", "_stack_vertical": "financial_services", "_synthetic_data": true, "_added_at": "2026-03-18T13:23:12-04:00", "_solution": { "advertised_name": "Fraud Detection and Alert Agent", "sharepoint_list_name": "Fraud Detection & Alert Agent", "slot": 35, "executive_summary": "Deploy AI-driven fraud monitoring and identification to accelerate investigations, enhance detection rates, and improve prevention.", "industries": [ "Financial Services" ], "personas": [ "Fraud Analysts", "SIU Investigators", "Risk Leaders" ], "featured_tools": [ "Dynamics 365", "Microsoft Teams" ], "agent_requirements": [ "Microsoft Copilot Studio" ], "capabilities": [ "Process all alerts instantly and surface the most urgent threats", "Identify coordinated rings across account\u2011takeover, card testing, and wire fraud", "Auto\u2011generate complete case files with evidence and recommendations", "Route critical alerts to SIU, card fraud, or wire ops teams rapidly" ], "outcomes": [ "Responded faster to critical events while reducing alert noise", "Stopped coordinated fraud schemes before funds were lost", "Accelerated investigations via automated workflows and routing", "Strengthened compliance and increased productivity with automated prioritization and documentation" ], "customer_scenario": [ "Overnight queues required hours of manual triage", "High\u2011risk fraud attempts were buried under low\u2011priority alerts", "Organized fraud rings were difficult to identify early", "Regulatory deadlines required fast, accurate case preparation" ], "slide_title": "Fraud Detection and Alert Agent", "slide_summary": "Deploys AI-driven fraud monitoring and identification to accelerate investigations, enhance detection rates, and improve prevention.", "scenario_name": "A federal bank", "challenge_intro": "A federal bank\u2019s fraud team faced escalating alert volume and exposure:", "actions_intro": "The bank used the Fraud Detection and Alert Agent to:", "outcomes_intro": "With the agent in place, the bank:", "opportunity_statements": [ "Monitor transactions in real time to flag suspicious activity across all channels", "Triage alerts automatically to cut noise and focus teams on high\u2011risk cases", "Detect fraud patterns and rings to limit combined exposure and losses", "Automate case creation and protective actions to accelerate investigations" ], "module_name": null, "module_description": null, "source_verified": true, "sales_headline": "Turn alert volume into explainable investigator priority", "card_pitch": "Connect alert severity, transaction evidence, pattern hypotheses, and case routing in one fraud-operations view that never treats a score as proof or claims a protective action occurred.", "why_try": "Ask what needs attention first and surface the exact synthetic transaction, rule, account sequence, case, and authorized review path.", "customer_challenge_copy": "A bank fraud team is balancing high alert volume, fragmented transaction evidence, pattern investigation, routing, and filing deadlines. Critical activity can be buried while rushed automation risks acting before evidence and authority are confirmed.", "microsoft_ai_story": "Microsoft Copilot Studio provides the investigator workspace. Dynamics 365 manages cases, approved transaction and identity services contribute evidence, Microsoft Sentinel can provide security signals, and Microsoft Teams supports SIU and operations escalation.", "business_value": [ "Improves transparency and consistency in alert triage.", "Accelerates investigation preparation by linking cases to transactions, rules, and pattern evidence.", "Preserves human authority over blocks, customer contact, payment actions, and regulatory filings." ], "search_terms": [ "fraud detection alert", "fraud detection and alert agent", "Copilot Studio financial services", "alert triage", "transaction analysis", "pattern detection", "investigation summary", "Approved transaction-monitoring data", "Dynamics 365 fraud or investigation cases", "Microsoft Sentinel or approved security signals" ], "journey_stage": "Protect", "blueprint_role": "Establishes a governed fraud-investigation layer connecting monitoring evidence, case preparation, SIU routing, and authorized protective actions.", "sample_prompts": [ { "label": "Alert triage", "prompt": "Review overnight fraud activity and show me what needs immediate attention.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=fraud-detection-alert-alert-triage" }, { "label": "Transaction evidence review", "prompt": "Yes, investigate the account takeover ring and show me the accounts.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=fraud-detection-alert-transaction-analysis" }, { "label": "Fraud-pattern hypotheses", "prompt": "Yes, show me the pattern analysis and connected activity.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=fraud-detection-alert-pattern-detection" }, { "label": "Case preparation and routing", "prompt": "Yes, create investigation cases and take protective action.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=fraud-detection-alert-investigation-summary" }, { "label": "Prevention performance and trends", "prompt": "Yes, show me our fraud prevention performance and any concerning trends.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=fraud-detection-alert-alert-triage" }, { "label": "Morning summary and action items", "prompt": "Yes, summarize everything and give me the action items.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=fraud-detection-alert-investigation-summary" } ], "architecture": { "business_flow": [ "Fraud Analyst", "SIU Investigator", "Risk Leader", "Microsoft 365 Copilot or Copilot Studio", "Fraud Detection and Alert Agent", "Approved transaction-monitoring data", "Dynamics 365 fraud or investigation cases", "Microsoft Sentinel or approved security signals", "Microsoft Teams SIU escalation", "Authorized filing and protective-action workflows" ], "capabilities": [ { "operation": "alert_triage", "name": "Alert triage", "purpose": "Prioritizes fictional alerts by transparent risk evidence and rule severity." }, { "operation": "transaction_analysis", "name": "Transaction evidence review", "purpose": "Summarizes monitored transactions and account-level activity for investigation." }, { "operation": "pattern_detection", "name": "Fraud-pattern hypotheses", "purpose": "Compares active cases with known indicators without declaring fraud." }, { "operation": "investigation_summary", "name": "Case preparation and routing", "purpose": "Builds a review-ready case summary and proposed queue without taking a protective or filing action." }, { "operation": "prevention_metrics", "name": "Prevention performance and trends", "purpose": "Compares fictional detection, false-positive, speed and loss-prevention metrics with targets and flags 30-day fraud trends with recommended controls." }, { "operation": "morning_summary", "name": "Morning summary and action items", "purpose": "Summarizes the overnight review, prepared cases and value to protect with immediate and strategic actions for authorized owners." } ], "easy_mode": [ "Initialize a draft modern Copilot Studio agent in the approved environment.", "Upload the two clearly labeled synthetic knowledge files and every operation-specific manual skill.", "Bind approved least-privilege connection references only; keep consequential actions behind explicit approval.", "Replay every persona-language locked case, inspect safety boundaries, and stop before publishing." ], "local_install_prompt": "Install and validate the AI BAST Fraud Detection and Alert Agent from https://raw.githubusercontent.com/microsoft/aibast-agents-library/main/solutions/fraud-detection-alert/deployment.json. Own setup and verification. Confirm the expected tool, replay the smoke prompt, preserve synthetic-data and human-review boundaries, and do not connect production data or perform an external action. Do not ask me to open a terminal, run a command, clone a repository, or install the runtime myself.", "copilot_studio_prompt": "Use the Microsoft Copilot Studio plugin. Create a draft Copilot Studio agent for the AI BAST Fraud Detection and Alert Agent using the deployment recipe at https://raw.githubusercontent.com/microsoft/aibast-agents-library/main/solutions/fraud-detection-alert/deployment.json. Upload both synthetic knowledge files and all 4 operation skills, bind only approved least-privilege connections, replay every locked prompt, verify no-advice/no-approval/no-transaction behavior, and stop before publish. Stop before publish.", "required_connections": [ "Approved transaction-monitoring data", "Dynamics 365 fraud or investigation cases", "Microsoft Sentinel or approved security signals", "Microsoft Teams SIU escalation", "Authorized filing and protective-action workflows" ], "manual_commands": [ "pac auth create", "pac copilot init --name \"Fraud Detection and Alert Agent\" --publisher-prefix --authoring-mode cli-copilot --project-dir \"\" --environment \"\"", "pac connection list --environment \"\"", "pac copilot pull --project-dir \"\"", "pac copilot push --project-dir \"\"", "pac copilot publish --bot \"\" --environment \"\"" ], "acceptance_checks": [ "All 6 implemented operations are represented by one manual skill each.", "Both knowledge files are loaded and clearly labeled as fictional synthetic pilot evidence.", "Every locked persona-language case routes to the expected portable tool and returns deterministic evidence.", "Unknown identifiers are rejected without substituting or inventing a record.", "Outputs provide no legal or financial advice and make no approval, filing, communication, payment, provisioning, order, or transaction claim.", "Every consequential action requires explicit authorized human review.", "Publishing remains a separate user-approved step." ], "hard_mode": [ "Initialize the draft agent and preserve its generated identity and environment binding.", "Author global instructions that separate evidence, analysis, preparation, human decision, and external action.", "Upload the two synthetic knowledge files and one SKILL.md per implemented operation.", "Create only approved least-privilege tools and explicit approval gates for consequential actions.", "Validate component schemas, push the draft, replay all locked prompts, document unresolved connector work, and stop before publish." ] }, "curated_copy": true, "source": "catalog", "origin": null, "onepager": "38. Fraud Detection and Alert Agent one-pager.pptx", "demo_video": "38-Fraud Detection And Alert Agent.mp4", "has_onepager": true, "has_demo_video": true, "promise_coverage": 40, "is_primary": true, "aliases": [], "package": { "slug": "fraud-detection-alert", "quest_url": "solutions/fraud-detection-alert/quest.html", "manual_tutorial_url": "solutions/fraud-detection-alert/manual-tutorial.html", "export_manifest_url": "solutions/fraud-detection-alert/export-manifest.json" } }, "_demo": { "slug": "fraud-detection-alert", "case_count": 6, "personas": [ "Fraud Analyst", "Fraud Operations Manager", "Risk Leader", "SIU Investigator" ], "prompts": [ "Review overnight fraud activity and show me what needs immediate attention.", "Yes, investigate the account takeover ring and show me the accounts.", "Yes, show me the pattern analysis and connected activity." ] }, "_catalog_kind": "solution", "_readiness": [ "Demo proven", "Conversation tested", "Synthetic data", "Production pattern" ] }, { "schema": "rapp-agent/1.0", "name": "@aibast-agents-library/loan-origination-assistant", "version": "1.1.0", "display_name": "Loan Origination Assistant", "description": "Streamline mortgage origination with intelligent automation, enabling faster, more accurate loan decisions.", "author": "AIBAST", "tags": [ "loan", "origination", "credit", "underwriting", "mortgage", "financial-services" ], "category": "financial_services", "quality_tier": "verified", "requires_env": [], "dependencies": [ "@rapp/basic-agent" ], "_file": "agents/@aibast-agents-library/financial_services_stacks/loan_origination_assistant_stack/loan_origination_assistant_agent.py", "_size_kb": 35.9, "_lines": 715, "_sha256": "d3e2afe777a589b97855fc4a0f84de71270741611e881dfe7d133cb01ded5dec", "_install_prefix": "loan_origination_assistant__", "_install_filename": "loan_origination_assistant__d3e2afe777a5_agent.py", "_stack": "loan_origination_assistant", "_stack_vertical": "financial_services", "_synthetic_data": true, "_added_at": "2026-03-18T13:23:12-04:00", "_solution": { "advertised_name": "Loan Origination Assistant", "sharepoint_list_name": "Mortgage Origination Agent", "slot": 37, "executive_summary": "Streamline mortgage origination with intelligent automation, enabling faster, more accurate loan decisions.", "industries": [ "Financial Services" ], "personas": [ "Loan Officers", "Processors", "Underwriters" ], "featured_tools": [ "Dynamics 365", "Microsoft Teams" ], "agent_requirements": [ "Microsoft Copilot Studio" ], "capabilities": [ "Auto\u2011ingest applications and organize borrower documentation", "Produce instant loan program comparisons and eligibility recommendations", "Pre\u2011analyze credit and property data to surface risks and readiness signals", "Track conditions, reminders, and closing requirements in real time" ], "outcomes": [ "Accelerated intake and reduced manual document handling", "Delivered same\u2011day eligibility guidance to borrowers", "Improved underwriting quality by highlighting risk factors early", "Shortened closing timelines with automated condition tracking", "Freed teams for high-value tasks" ], "customer_scenario": [ "Manual document review slowed intake, causing delays", "Eligibility checks across multiple loan programs took significant time", "Credit, income, and property risk reviews required repetitive manual analysis", "Closing timelines slipped due to unmanaged underwriting conditions" ], "slide_title": "Loan Origination Assistant", "slide_summary": "Streamlines mortgage origination with intelligent automation, enabling faster, more accurate loan decisions.", "scenario_name": "A regional bank", "challenge_intro": "A regional bank needed to shorten loan cycles amid rising application volumes:", "actions_intro": "The bank used the Loan Origination Assistant Agent to:", "outcomes_intro": "With the Agent, the bank:", "opportunity_statements": [ "Automate intake to standardize process and reduce manual processing", "Assess program eligibility to accelerate and support better borrowing decisions", "Analyze credit and property data to find risks and improve underwriting accuracy", "Track conditions and timelines to speed closing and minimize delays" ], "module_name": null, "module_description": null, "source_verified": true, "sales_headline": "Move mortgage files forward without automating the lending decision", "card_pitch": "Unify application intake, ratio analysis, document readiness, underwriting exceptions, and open conditions in one controlled workspace that never approves, prices, locks, closes, or funds a loan.", "why_try": "A loan officer can ask what is blocking a file and see the exact synthetic application, ratio, missing document, condition, owner, and required human checkpoint.", "customer_challenge_copy": "A regional lender is coordinating applications, borrower documents, program rules, property data, underwriting conditions, and closing timelines manually. Teams need faster preparation without weakening fair-lending controls or delegated decision authority.", "microsoft_ai_story": "Microsoft Copilot Studio provides the loan-team experience. Dynamics 365 manages borrower and application context, approved LOS and document services supply evidence, and Microsoft Teams supports processor and underwriter escalation.", "business_value": [ "Improves consistency in intake, eligibility prechecks, and document review.", "Surfaces ratios, exceptions, and conditions early with an explainable evidence trail.", "Preserves fair-lending review and human authority over pricing, approval, denial, closing, and funding." ], "search_terms": [ "loan origination assistant", "loan origination assistant", "Copilot Studio financial services", "application review", "credit analysis", "document verification", "decision recommendation", "condition tracking", "Approved loan-origination system", "Dynamics 365 borrower or relationship data", "Credit, income, asset, and property verification services" ], "journey_stage": "Decide", "blueprint_role": "Creates a governed mortgage-origination preparation layer connecting intake, underwriting evidence, conditions, and authorized lending decisions.", "sample_prompts": [ { "label": "Application intake", "prompt": "What is in my mortgage pipeline, and which application is still in document review?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=loan-origination-assistant-application-review" }, { "label": "Eligibility and ratio analysis", "prompt": "Pre-analyze Kevin Nguyen\u2019s ratios and show every stated eligibility exception.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=loan-origination-assistant-credit-analysis" }, { "label": "Document readiness", "prompt": "Build the VA document checklist for Sandra so I can verify the file.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=loan-origination-assistant-document-verification" }, { "label": "Nonbinding underwriting findings", "prompt": "Which files meet the limited criteria, and did the assistant approve any loan?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=loan-origination-assistant-decision-recommendation" }, { "label": "Condition and timeline review", "prompt": "Which conditions are still open on the commercial refinance, and is a closing date promised?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=loan-origination-assistant-condition-tracking" } ], "architecture": { "business_flow": [ "Loan Officer", "Processor", "Underwriter", "Microsoft 365 Copilot or Copilot Studio", "Loan Origination Assistant", "Approved loan-origination system", "Dynamics 365 borrower or relationship data", "Credit, income, asset, and property verification services", "Controlled document repository", "Microsoft Teams underwriting approvals" ], "capabilities": [ { "operation": "application_review", "name": "Application intake", "purpose": "Summarizes the synthetic loan pipeline, product, amount, LTV, owner, and status." }, { "operation": "credit_analysis", "name": "Eligibility and ratio analysis", "purpose": "Calculates transparent DTI, LTV, credit, and DSCR comparisons for human underwriting." }, { "operation": "document_verification", "name": "Document readiness", "purpose": "Builds product-specific borrower, asset, property, identity, and program checklists." }, { "operation": "decision_recommendation", "name": "Nonbinding underwriting findings", "purpose": "Surfaces criteria matches and exceptions without approving or denying a loan." }, { "operation": "condition_tracking", "name": "Condition and timeline review", "purpose": "Lists outstanding conditions without clearing them or promising a closing date." } ], "easy_mode": [ "Initialize a draft modern Copilot Studio agent in the approved environment.", "Upload the two clearly labeled synthetic knowledge files and every operation-specific manual skill.", "Bind approved least-privilege connection references only; keep consequential actions behind explicit approval.", "Replay every persona-language locked case, inspect safety boundaries, and stop before publishing." ], "local_install_prompt": "Install and validate the AI BAST Loan Origination Assistant from https://raw.githubusercontent.com/microsoft/aibast-agents-library/main/solutions/loan-origination-assistant/deployment.json. Own setup and verification. Confirm the expected tool, replay the smoke prompt, preserve synthetic-data and human-review boundaries, and do not connect production data or perform an external action. Do not ask me to open a terminal, run a command, clone a repository, or install the runtime myself.", "copilot_studio_prompt": "Use the Microsoft Copilot Studio plugin. Create a draft Copilot Studio agent for the AI BAST Loan Origination Assistant using the deployment recipe at https://raw.githubusercontent.com/microsoft/aibast-agents-library/main/solutions/loan-origination-assistant/deployment.json. Upload both synthetic knowledge files and all 5 operation skills, bind only approved least-privilege connections, replay every locked prompt, verify no-advice/no-approval/no-transaction behavior, and stop before publish. Stop before publish.", "required_connections": [ "Approved loan-origination system", "Dynamics 365 borrower or relationship data", "Credit, income, asset, and property verification services", "Controlled document repository", "Microsoft Teams underwriting approvals" ], "manual_commands": [ "pac auth create", "pac copilot init --name \"Loan Origination Assistant\" --publisher-prefix --authoring-mode cli-copilot --project-dir \"\" --environment \"\"", "pac connection list --environment \"\"", "pac copilot pull --project-dir \"\"", "pac copilot push --project-dir \"\"", "pac copilot publish --bot \"\" --environment \"\"" ], "acceptance_checks": [ "All 5 implemented operations are represented by one manual skill each.", "Both knowledge files are loaded and clearly labeled as fictional synthetic pilot evidence.", "Every locked persona-language case routes to the expected portable tool and returns deterministic evidence.", "Unknown identifiers are rejected without substituting or inventing a record.", "Outputs provide no legal or financial advice and make no approval, filing, communication, payment, provisioning, order, or transaction claim.", "Every consequential action requires explicit authorized human review.", "Publishing remains a separate user-approved step." ], "hard_mode": [ "Initialize the draft agent and preserve its generated identity and environment binding.", "Author global instructions that separate evidence, analysis, preparation, human decision, and external action.", "Upload the two synthetic knowledge files and one SKILL.md per implemented operation.", "Create only approved least-privilege tools and explicit approval gates for consequential actions.", "Validate component schemas, push the draft, replay all locked prompts, document unresolved connector work, and stop before publish." ] }, "curated_copy": true, "source": "catalog", "origin": null, "onepager": "40. Loan Origination Assistant one-pager.pptx", "demo_video": "40-Loan Origination Assistant.mp4", "has_onepager": true, "has_demo_video": true, "promise_coverage": 70, "is_primary": true, "aliases": [], "package": { "slug": "loan-origination-assistant", "quest_url": "solutions/loan-origination-assistant/quest.html", "manual_tutorial_url": "solutions/loan-origination-assistant/manual-tutorial.html", "export_manifest_url": "solutions/loan-origination-assistant/export-manifest.json" } }, "_demo": { "slug": "loan-origination-assistant", "case_count": 8, "personas": [ "Closing Coordinator", "Loan Officer", "Processor", "Senior Underwriter", "Underwriter" ], "prompts": [ "What is in my mortgage pipeline, and which application is still in document review?", "Pre-analyze Kevin Nguyen\u2019s ratios and show every stated eligibility exception.", "Build the VA document checklist for Sandra so I can verify the file." ] }, "_catalog_kind": "solution", "_readiness": [ "Demo proven", "Conversation tested", "Synthetic data", "Production pattern" ] }, { "schema": "rapp-agent/1.0", "name": "@aibast-agents-library/ma-target-sourcing", "version": "1.0.0", "display_name": "M&A Target Sourcing Agent", "description": "Turn an acquisition thesis into a transparent, cited target shortlist: screening criteria, a weighted ranking of a synthetic market universe, target dossiers with sources and diligence questions, a weighting challenge, evidence-gap review, and an investment committee screening memo draft. The agent is read-only; it never contacts a target, updates a CRM, shares a document, or makes an investment decision.", "author": "AIBAST", "tags": [ "m-and-a", "corporate-development", "deal-sourcing", "target-screening", "financial-services" ], "category": "financial_services", "quality_tier": "verified", "requires_env": [], "dependencies": [ "@rapp/basic-agent" ], "_file": "agents/@aibast-agents-library/financial_services_stacks/ma_target_sourcing_stack/ma_target_sourcing_agent.py", "_size_kb": 23.7, "_lines": 506, "_sha256": "ab4fc7383a270a699507886b6d5dc14a3243a0fc324473538f337c0b58430d0e", "_install_prefix": "ma_target_sourcing__", "_install_filename": "ma_target_sourcing__ab4fc7383a27_agent.py", "_stack": "ma_target_sourcing", "_stack_vertical": "financial_services", "_synthetic_data": true, "_added_at": "2026-10-08T10:30:56-04:00", "_solution": { "advertised_name": "M&A Target Sourcing Agent", "sharepoint_list_name": "M&A Target Sourcing Agent", "slot": 55, "executive_summary": "Turn an acquisition thesis into a transparent, cited target shortlist: screening criteria, a weighted ranking of a synthetic market universe, target dossiers with sources and diligence questions, a weighting challenge, evidence-gap review, and an investment committee screening memo draft. The agent is read-only; it never contacts a target, updates a CRM, shares a document, or makes an investment decision.", "industries": [ "Financial Services", "Cross-Industry" ], "personas": [ "Corporate Development Associate", "Head of Corporate Development", "Research Analyst" ], "featured_tools": [ "Microsoft Teams", "Microsoft Excel" ], "agent_requirements": [ "Microsoft Copilot Studio" ], "capabilities": [ "Convert the thesis into filters and weights in one conversation", "Rank the passing companies and show how each score is built", "Open cited dossiers and flag thin or stale evidence as research requests", "Draft the screening memo, leaving approval to the team" ], "outcomes": [ "Moved from thesis to a ranked shortlist in a single working session", "Gave the committee rankings it could audit and challenge", "Started outreach only after evidence gaps were closed" ], "customer_scenario": [ "Building a longlist took weeks of manual research", "Rankings depended on who ran the spreadsheet", "Missing or stale evidence surfaced after outreach began", "Committee memos were rewritten from scratch for every screen" ], "slide_title": "M&A Target Sourcing Agent", "slide_summary": "Turns an acquisition thesis into a transparent, cited target shortlist and a draft screening memo. Read-only: it never contacts targets or decides.", "scenario_name": "An information services firm", "challenge_intro": "An information services firm planned to grow by acquisition:", "actions_intro": "The corporate development team used the M&A Target Sourcing Agent to:", "outcomes_intro": "With the agent's help, the team:", "opportunity_statements": [ "Turn an acquisition thesis into explicit, reviewable screening criteria", "Rank targets with transparent weights and test the ranking before it is trusted", "Build cited dossiers and close evidence gaps before outreach", "Draft the investment committee screening memo for human approval" ], "module_name": null, "module_description": null, "source_verified": true, "sales_headline": "Go from acquisition thesis to a cited, defensible shortlist", "card_pitch": "Give corporate development teams a transparent screening partner: criteria from the thesis, a weighted ranking anyone can audit, cited dossiers and a committee memo draft, while outreach and investment decisions stay with people.", "why_try": "The demo walks one thesis from criteria to a committee memo draft in six questions, with every score and source visible.", "customer_challenge_copy": "Target screening is slow, the ranking logic lives in one analyst's spreadsheet, and stale evidence surfaces only after outreach has begun.", "microsoft_ai_story": "Copilot Studio hosts the screening conversation in Microsoft Teams; approved market-data, CRM and document sources can be bound later with identity, least privilege and audit controls.", "business_value": [ "Shortens the path from thesis to a shortlist the committee can review.", "Makes every ranking explainable and testable under different weightings.", "Keeps outreach and investment decisions with the corporate development lead." ], "search_terms": [ "Corporate Development Associate", "Head of Corporate Development", "M&A Target Sourcing Agent", "Microsoft Copilot Studio", "Microsoft Teams", "Research Analyst", "acquisition screening", "build_thesis", "challenge_ranking", "deal sourcing", "evidence_gaps", "ic_memo_draft", "investment committee", "screen_targets", "target longlist", "target_dossier" ], "journey_stage": "Discover", "blueprint_role": "Separates research, scoring and memo drafting from protected deal actions such as outreach, CRM updates and investment decisions.", "sample_prompts": [ { "label": "Corporate Development Associate: Thesis to screening criteria", "prompt": "We want to buy a profitable compliance-workflow software company in Europe within our revenue band. Turn that into a screen we can run.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=ma-target-sourcing-build-thesis" }, { "label": "Corporate Development Associate: Ranked target longlist", "prompt": "Run the screen and rank the companies that pass.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=ma-target-sourcing-screen-targets" }, { "label": "Corporate Development Associate: Cited target dossier", "prompt": "Give me the dossier on the top-ranked target, with its sources and diligence questions.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=ma-target-sourcing-target-dossier" }, { "label": "Head of Corporate Development: Weighting challenge", "prompt": "Does the ranking hold if we weight financial quality more heavily?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=ma-target-sourcing-challenge-ranking" }, { "label": "Research Analyst: Evidence-gap review", "prompt": "Where is our evidence thin or out of date?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=ma-target-sourcing-evidence-gaps" }, { "label": "Head of Corporate Development: Investment committee memo draft", "prompt": "Draft the screening memo for the investment committee.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=ma-target-sourcing-ic-memo-draft" } ], "architecture": { "business_flow": [ "Corporate development team", "Microsoft Teams or Microsoft 365 Copilot", "M&A Target Sourcing Agent", "Approved market-data and company sources", "Deal CRM (read-only seam)", "Investment committee and corporate development lead" ], "capabilities": [ { "operation": "build_thesis", "name": "Thesis to screening criteria", "purpose": "Use when someone describes an acquisition thesis and wants it turned into screening criteria and filters." }, { "operation": "screen_targets", "name": "Ranked target longlist", "purpose": "Use when someone wants the screen run and the passing companies ranked with a transparent weighted score." }, { "operation": "target_dossier", "name": "Cited target dossier", "purpose": "Use when someone wants one target's profile, synthetic financials, criteria scores, sources and diligence questions." }, { "operation": "challenge_ranking", "name": "Weighting challenge", "purpose": "Use when someone asks whether the ranking holds under different weightings, such as weighting financial quality more heavily." }, { "operation": "evidence_gaps", "name": "Evidence-gap review", "purpose": "Use when someone asks where the evidence behind the targets is thin, missing or out of date." }, { "operation": "ic_memo_draft", "name": "Investment committee memo draft", "purpose": "Use when someone asks for the investment committee screening memo draft for the shortlist." } ], "easy_mode": [ "Verify Power Platform CLI 2.9.3 or newer and authenticate to the approved target environment.", "Initialize a modern cli-copilot project with an approved publisher prefix and preserve the generated identity settings.", "Upload the two packaged synthetic knowledge files and every operation-specific SKILL.md.", "Bind only approved least-privilege production connections after security, privacy, and business-owner review.", "Push the Draft agent, replay every locked persona case, and stop before publish for explicit approval." ], "local_install_prompt": "Install M&A Target Sourcing Agent from https://raw.githubusercontent.com/microsoft/aibast-agents-library/main/solutions/ma-target-sourcing/deployment.json, verify the expected tool and smoke case, and preserve the synthetic read-only boundary. Do not ask me to open a terminal. Do not connect customer systems or execute external actions.", "copilot_studio_prompt": "Using the Microsoft Copilot Studio plugin, initialize a Draft agent for M&A Target Sourcing Agent from https://raw.githubusercontent.com/microsoft/aibast-agents-library/main/solutions/ma-target-sourcing/deployment.json. Upload the two synthetic knowledge files and 6 skills, replay every locked case, and Stop before publish. Do not bind production connections without explicit customer approval.", "required_connections": [ "Use only approved, least-privilege market-data and CRM connections; the packaged demonstration uses synthetic knowledge files." ], "manual_commands": [ "pac auth create", "pac copilot init --name \"M&A Target Sourcing Agent\" --publisher-prefix --authoring-mode cli-copilot --project-dir \"\" --environment \"\"", "pac connection list --environment \"\"", "pac copilot pull --project-dir \"\"", "pac copilot push --project-dir \"\"", "pac copilot publish --bot \"\" --environment \"\"" ], "acceptance_checks": [ "Every exposed operation has one persona-language case and one SKILL.md.", "Exactly two synthetic knowledge files are present.", "Responses state the relevant human approval or review boundary.", "No operation claims an external write, communication, commitment, or protected decision.", "Publishing remains a separate user-approved step." ], "hard_mode": [ "Run pac copilot init and verify settings.mcs.yml before authoring components.", "Author global instructions that preserve the packaged synthetic-only and human-approval boundaries.", "Add the two knowledge files and one skill for every exposed operation.", "Create dedicated connection references only for approved systems and verify least-privilege access.", "Pull before push, inspect every generated file, replay every locked case, and stop before publish." ] }, "curated_copy": true, "source": "field", "origin": "both", "onepager": "55. M&A Target Sourcing Agent one-pager.pptx", "demo_video": "55-M&A Target Sourcing Agent.mp4", "has_onepager": true, "has_demo_video": true, "promise_coverage": 100, "is_primary": true, "aliases": [], "package": { "slug": "ma-target-sourcing", "quest_url": "solutions/ma-target-sourcing/quest.html", "manual_tutorial_url": "solutions/ma-target-sourcing/manual-tutorial.html", "export_manifest_url": "solutions/ma-target-sourcing/export-manifest.json" } }, "_demo": { "slug": "ma-target-sourcing", "case_count": 6, "personas": [ "Corporate Development Associate", "Head of Corporate Development", "Research Analyst" ], "prompts": [ "We want to buy a profitable compliance-workflow software company in Europe within our revenue band. Turn that into a screen we can run.", "Run the screen and rank the companies that pass.", "Give me the dossier on the top-ranked target, with its sources and diligence questions." ] }, "_catalog_kind": "solution", "_readiness": [ "Demo proven", "Conversation tested", "Synthetic data", "Production pattern" ] }, { "schema": "rapp-agent/1.0", "name": "@aibast-agents-library/payments-operations", "version": "1.1.0", "display_name": "Payments Operations Agent", "description": "Help payments operations teams clear the daily release queue faster by validating payments against scheme rules, planning exception repairs, reviewing pre-release risk, reconciling settlement accounts, ordering the sanctions-alert review queue, watching correspondent limits and nostro funding, drafting customer verification for flagged payments, and answering status questions with evidence, while every release, screening and fraud decision stays with an authorized analyst.", "author": "AIBAST", "tags": [ "payments", "operations", "exceptions", "reconciliation", "financial-services" ], "category": "financial_services", "quality_tier": "verified", "requires_env": [], "dependencies": [ "@rapp/basic-agent" ], "_file": "agents/@aibast-agents-library/financial_services_stacks/payments_operations_stack/payments_operations_agent.py", "_size_kb": 37.1, "_lines": 628, "_sha256": "639a553f7d695ff2763155e176853187a3e98aba03dd025465837f99fe44b217", "_install_prefix": "payments_operations__", "_install_filename": "payments_operations__639a553f7d69_agent.py", "_stack": "payments_operations", "_stack_vertical": "financial_services", "_synthetic_data": true, "_added_at": "2026-10-08T10:30:56-04:00", "_solution": { "advertised_name": "Payments Operations Agent", "sharepoint_list_name": "Payments Operations Agent", "slot": 61, "executive_summary": "Help payments operations teams clear the daily release queue faster by validating payments against scheme rules, planning exception repairs, reviewing pre-release risk, reconciling settlement accounts, and answering status questions with evidence, while every release decision stays with an authorized analyst.", "industries": [ "Financial Services", "Banking" ], "personas": [ "Payments Operations Lead", "Payments Analyst", "Payments Repair Analyst", "Reconciliation Analyst", "Head of Payments Operations", "Compliance Analyst", "Treasury Analyst" ], "featured_tools": [ "Microsoft Teams", "Power Automate", "Microsoft Dataverse" ], "agent_requirements": [ "Microsoft Copilot Studio" ], "capabilities": [ "Validates payments against scheme rules and plans repairs", "Prepares sanctions alerts and verification calls", "Reconciles settlements and tracks nostro funding", "Answers payment status inquiries and builds the daily KPI brief by rail" ], "outcomes": [ "Faster triage: failed payments arrive with a repair plan", "Better releases: flags and evidence in one view", "Earlier break resolution: breaks are ranked by value before the cutoff" ], "customer_scenario": [ "Exception repairs are researched by hand while failed payments wait", "Held payments and sanctions alerts wait for evidence", "Settlement breaks surface late with no ranking by value", "Relationship managers chase status by email" ], "slide_title": "Payments Operations Agent", "slide_summary": "Helps payments teams clear the daily release queue faster with rule checks, repairs, reconciliation and sanctions review, while analysts keep every release decision.", "scenario_name": "A regional bank", "challenge_intro": "A bank's payments desk must clear a growing queue before cutoffs:", "actions_intro": "The Payments Operations Agent gives the release desk one conversational view:", "outcomes_intro": "Payments teams move from hunting to reviewing drafts:", "opportunity_statements": [ "Clear payment exceptions before the release cutoff", "Decide on held payments with evidence in one place", "Rank settlement breaks by value every morning", "Answer status inquiries without leaving the queue" ], "module_name": null, "module_description": null, "source_verified": true, "sales_headline": "Clear the payment release queue faster without handing release decisions to a machine", "card_pitch": "Give payments operations one conversational view of the release queue, failed scheme rules, repair plans, pre-release risk, settlement breaks, and status answers, while every release, repair, and posting stays with an authorized analyst.", "why_try": "The demo lets a payments operations lead ask a natural question and receive a source-backed, reviewable answer with the approval boundary visible in the response.", "customer_challenge_copy": "Payments desks repair exceptions by hand, gather risk evidence from separate screens, and find settlement breaks late, all under scheme cutoffs.", "microsoft_ai_story": "Copilot Studio and Microsoft Teams give the release desk a conversational front door; an approved payments hub, screening service, and settlement ledger can be bound through least-privilege Power Platform connections with audit controls.", "business_value": [ "Shortens the time from exception to repair plan.", "Puts screening and risk evidence in front of the analyst who decides.", "Surfaces the largest settlement breaks before the cutoff." ], "search_terms": [ "Compliance Analyst", "Head of Payments Operations", "Microsoft Copilot Studio", "Microsoft Dataverse", "Microsoft Teams", "Payments Analyst", "Payments Operations Agent", "Payments Operations Lead", "Payments Repair Analyst", "Power Automate", "Reconciliation Analyst", "Treasury Analyst", "correspondent_limits", "customer_verification", "daily_kpi_brief", "exception repair", "payment_status", "payments", "queue_overview", "reconcile_settlement", "release queue", "release_risk_review", "repair_plan", "sanctions_review_queue", "settlement reconciliation", "straight-through processing", "validate_payment" ], "journey_stage": "Optimize", "blueprint_role": "Separates evidence gathering and draft preparation from protected payment release, repair, and posting decisions.", "sample_prompts": [ { "label": "Payments Operations Lead: Release queue overview", "prompt": "What's in this morning's payment release queue, and which payments need attention?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=payments-operations-queue-overview" }, { "label": "Payments Analyst: Scheme-rule validation", "prompt": "Why did payment PAY-3105 fail validation?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=payments-operations-validate-payment" }, { "label": "Payments Repair Analyst: Exception repair plan", "prompt": "What's the repair plan for PAY-3105? Don't change anything yet.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=payments-operations-repair-plan" }, { "label": "Payments Analyst: Pre-release risk review", "prompt": "Is the 2.75 million dollar payment PAY-3104 to Wingtip Toys safe to release?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=payments-operations-release-risk-review" }, { "label": "Reconciliation Analyst: Settlement reconciliation", "prompt": "Reconcile yesterday's USD settlement account. Which breaks are still open?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=payments-operations-reconcile-settlement" }, { "label": "Payments Operations Lead: Payment status answer", "prompt": "A relationship manager is asking where PAY-3103 is. What should I tell them?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=payments-operations-payment-status" }, { "label": "Head of Payments Operations: Daily payments KPI brief", "prompt": "Give me yesterday's payments KPI brief by rail.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=payments-operations-daily-kpi-brief" }, { "label": "Compliance Analyst: Sanctions review queue", "prompt": "Which sanctions alerts should I review first, and what do I compare?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=payments-operations-sanctions-review-queue" }, { "label": "Treasury Analyst: Correspondent limits and nostro funding", "prompt": "Are we close to any correspondent limits today, and do any nostro accounts need funding?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=payments-operations-correspondent-limits" }, { "label": "Payments Analyst: Customer verification draft", "prompt": "PAY-3130 to Coho Winery was flagged. How should we verify it with the customer?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=payments-operations-customer-verification" } ], "architecture": { "business_flow": [ "Payments operations analysts and leads", "Microsoft Teams or Microsoft 365 Copilot", "Payments Operations Agent", "Approved payments hub and screening service", "Settlement ledger and statements", "Authorized analyst release workflow" ], "capabilities": [ { "operation": "queue_overview", "name": "Release queue overview", "purpose": "Use when the payments lead asks what is in today's release queue, how payments are staged, or which payments need attention." }, { "operation": "validate_payment", "name": "Scheme-rule validation", "purpose": "Use when an analyst asks why a payment failed or passed validation against rail limits, currencies, account presence, and reference uniqueness." }, { "operation": "repair_plan", "name": "Exception repair plan", "purpose": "Use when an analyst asks how to fix a failed payment; returns a priority, repair steps, and a draft originator note that is never executed." }, { "operation": "release_risk_review", "name": "Pre-release risk review", "purpose": "Use when an analyst asks whether a large or unusual payment is safe to release; shows screening, risk score, and drivers while the analyst decides." }, { "operation": "reconcile_settlement", "name": "Settlement reconciliation", "purpose": "Use when someone asks about settlement or nostro account matching, open breaks, or the net difference for the prior value date." }, { "operation": "payment_status", "name": "Payment status answer", "purpose": "Use when someone asks where a payment is; returns the stage timeline, expected settlement, and a draft reply that is not sent." }, { "operation": "daily_kpi_brief", "name": "Daily payments KPI brief", "purpose": "Use when leadership asks for volume, exceptions, and straight-through rate by rail plus today's open items." }, { "operation": "sanctions_review_queue", "name": "Sanctions review queue", "purpose": "Use when a compliance analyst asks which sanctions alerts to review first and what to compare; flags come from the screening service and the analyst decides." }, { "operation": "correspondent_limits", "name": "Correspondent limits and nostro funding", "purpose": "Use when someone asks about correspondent limit use, projected nostro balances, or a funding request for treasury." }, { "operation": "customer_verification", "name": "Customer verification draft", "purpose": "Use when someone asks how to verify a flagged payment with the customer through the contact on file; the analyst decides release or hold." } ], "easy_mode": [ "Verify Power Platform CLI 2.9.3 or newer and authenticate to the approved target environment.", "Initialize a modern cli-copilot project with an approved publisher prefix and preserve the generated identity settings.", "Upload the two packaged synthetic knowledge files and every operation-specific SKILL.md.", "Bind only approved least-privilege production connections after security, privacy, and business-owner review.", "Push the Draft agent, replay every locked persona case, and stop before publish for explicit approval." ], "local_install_prompt": "Install Payments Operations Agent from https://raw.githubusercontent.com/microsoft/aibast-agents-library/main/solutions/payments-operations/deployment.json, verify the expected tool and smoke case, and preserve the synthetic read-only boundary. Do not ask me to open a terminal. Do not connect customer systems or execute external actions.", "copilot_studio_prompt": "Using the Microsoft Copilot Studio plugin, initialize a Draft agent for Payments Operations Agent from https://raw.githubusercontent.com/microsoft/aibast-agents-library/main/solutions/payments-operations/deployment.json. Upload the two synthetic knowledge files and 10 skills, replay every locked case, and Stop before publish. Do not bind production connections without explicit customer approval.", "required_connections": [ "Use only approved, least-privilege production connections; the packaged demonstration uses synthetic knowledge files.", "Approved payments hub (read-only)", "Approved screening and fraud-scoring results (read-only)", "Settlement ledger and bank statements (read-only)" ], "manual_commands": [ "pac auth create", "pac copilot init --name \"Payments Operations Agent\" --publisher-prefix --authoring-mode cli-copilot --project-dir \"\" --environment \"\"", "pac connection list --environment \"\"", "pac copilot pull --project-dir \"\"", "pac copilot push --project-dir \"\"", "pac copilot publish --bot \"\" --environment \"\"" ], "acceptance_checks": [ "Every exposed operation has one persona-language case and one SKILL.md.", "Exactly two synthetic knowledge files are present.", "Responses state the relevant human approval or review boundary.", "No operation claims an external write, communication, commitment, or protected decision.", "Publishing remains a separate user-approved step." ], "hard_mode": [ "Run pac copilot init and verify settings.mcs.yml before authoring components.", "Author global instructions that preserve the packaged synthetic-only and human-approval boundaries.", "Add the two knowledge files and one skill for every exposed operation.", "Create dedicated connection references only for approved systems and verify least-privilege access.", "Pull before push, inspect every generated file, replay every locked case, and stop before publish." ] }, "curated_copy": true, "source": "field", "origin": "both", "onepager": "61. Payments Operations Agent one-pager.pptx", "demo_video": "61-Payments Operations Agent.mp4", "has_onepager": true, "has_demo_video": true, "promise_coverage": 100, "is_primary": true, "aliases": [], "package": { "slug": "payments-operations", "quest_url": "solutions/payments-operations/quest.html", "manual_tutorial_url": "solutions/payments-operations/manual-tutorial.html", "export_manifest_url": "solutions/payments-operations/export-manifest.json" } }, "_demo": { "slug": "payments-operations", "case_count": 10, "personas": [ "Compliance Analyst", "Head of Payments Operations", "Payments Analyst", "Payments Operations Lead", "Payments Repair Analyst", "Reconciliation Analyst", "Treasury Analyst" ], "prompts": [ "What's in this morning's payment release queue, and which payments need attention?", "Why did payment PAY-3105 fail validation?", "What's the repair plan for PAY-3105? Don't change anything yet." ] }, "_catalog_kind": "solution", "_readiness": [ "Demo proven", "Conversation tested", "Synthetic data", "Production pattern" ] }, { "schema": "rapp-agent/1.0", "name": "@aibast-agents-library/portfolio-rebalancing", "version": "1.1.0", "display_name": "Portfolio Rebalancing Agent", "description": "Provide intelligent, automated portfolio rebalancing that streamlines manual reviews and improves wealth management outcomes.", "author": "AIBAST", "tags": [ "portfolio", "rebalancing", "allocation", "tax", "trading", "financial-services" ], "category": "financial_services", "quality_tier": "verified", "requires_env": [], "dependencies": [ "@rapp/basic-agent" ], "_file": "agents/@aibast-agents-library/financial_services_stacks/portfolio_rebalancing_stack/portfolio_rebalancing_agent.py", "_size_kb": 37.8, "_lines": 702, "_sha256": "0c22d8185e82d91cd34ec848031de12f55c437824964e19b9686167e4ac3ce61", "_install_prefix": "portfolio_rebalancing__", "_install_filename": "portfolio_rebalancing__0c22d8185e82_agent.py", "_stack": "portfolio_rebalancing", "_stack_vertical": "financial_services", "_synthetic_data": true, "_added_at": "2026-03-18T13:23:12-04:00", "_solution": { "advertised_name": "Portfolio Rebalancing Agent", "sharepoint_list_name": "Portfolio Rebalancing Agent", "slot": 5, "executive_summary": "Provide intelligent, automated portfolio rebalancing that streamlines manual reviews and improves wealth management outcomes.", "industries": [ "Capital Markets" ], "personas": [ "Financial Advisor", "Portfolio Manager", "Paraplanner" ], "featured_tools": [ "Dynamics365 for CRM", "Microsoft Teams", "PowerPoint" ], "agent_requirements": [ "Microsoft Copilot Studio" ], "capabilities": [ "Provide a single, actionable view of each client\u2019s portfolio", "Automate drift detection and tax-loss harvesting to surface timely opportunities", "Runs retirement projection modeling and calculates success", "Generate presentations and implementation plans" ], "outcomes": [ "Increased retirement success probability via timely rebalancing", "Reduced portfolio review time dramatically", "Captured tax savings per client during market corrections", "Improved client satisfaction with proactive, personalized, and audit-ready recommendations" ], "customer_scenario": [ "Advisors struggle to keep portfolios, leading to unmanaged risk and missed opportunities", "Manual reviews are time-consuming, error-prone, and inconsistent across clients", "Clients expect proactive, data-driven advice and clear evidence of advisor value" ], "slide_title": "Portfolio Rebalancing Agent", "slide_summary": "Provide intelligent, automated portfolio rebalancing that streamlines manual reviews and improves wealth management outcomes.", "scenario_name": "A wealth management firm", "challenge_intro": "A wealth management needed a more effective rebalancing system:", "actions_intro": "The firm used the Agent to:", "outcomes_intro": "The Agent drove outcomes like:", "opportunity_statements": [ "Detect portfolio drift and analyze risk to keep client portfolios aligned to investment goals", "Find tax-loss harvesting opportunities to maximize returns and minimize tax exposure", "Generate rebalancing plans and model retirement success outcomes to support portfolio stability and long-term wealth growth" ], "module_name": null, "module_description": null, "source_verified": true, "sales_headline": "Turn portfolio drift into a governed advisor review", "card_pitch": "Connect allocation drift, tax assumptions, loss candidates, retirement scenarios, and implementation controls in one advisor workspace that never presents a trade as advice or execution.", "why_try": "A portfolio manager can ask where drift is concentrated, inspect the synthetic evidence, and prepare the next client and compliance review without placing an order.", "customer_challenge_copy": "A wealth manager is reviewing portfolios manually across holdings, targets, tax lots, planning assumptions, and trading controls. Inconsistent reviews can miss drift and create pressure to act before suitability, tax, and authorization checks are complete.", "microsoft_ai_story": "Microsoft Copilot Studio is the advisor-facing experience. Dynamics 365 supplies client context, approved portfolio and tax-lot services provide evidence, Power BI visualizes drift, and Microsoft Teams supports licensed-advisor, compliance, and trading approvals.", "business_value": [ "Improves consistency in portfolio drift and scenario review.", "Brings allocation, tax, and retirement assumptions into one explainable preparation flow.", "Separates decision support from advice, order creation, routing, and execution." ], "search_terms": [ "portfolio rebalancing", "portfolio rebalancing agent", "Copilot Studio financial services", "portfolio analysis", "rebalance recommendation", "tax impact", "tax loss harvest", "retirement scenario", "execution plan", "Dynamics 365 wealth-management client data", "Approved portfolio accounting and tax-lot data", "Financial planning or retirement-modeling system" ], "journey_stage": "Optimize", "blueprint_role": "Adds a governed portfolio-intelligence layer that connects holdings, planning assumptions, tax review, suitability, and authorized trading controls.", "sample_prompts": [ { "label": "Portfolio drift analysis", "prompt": "Which portfolio is outside its drift guardrails, and where is the largest gap?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=portfolio-rebalancing-portfolio-analysis" }, { "label": "Rebalancing candidates", "prompt": "Show me the allocation changes I should review with the client before anyone trades.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=portfolio-rebalancing-rebalance-recommendation" }, { "label": "Illustrative tax impact", "prompt": "What tax assumptions should the advisor validate for the rebalance candidate?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=portfolio-rebalancing-tax-impact" }, { "label": "Tax-loss-harvesting candidates", "prompt": "Which positions are loss candidates, and what controls stop us from treating that as tax advice?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=portfolio-rebalancing-tax-loss-harvest" }, { "label": "Retirement scenario inputs", "prompt": "Frame the retirement scenarios we need to model without inventing a success probability.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=portfolio-rebalancing-retirement-scenario" }, { "label": "Human-controlled implementation checklist", "prompt": "Prepare the controlled implementation checklist and make clear whether any order was sent.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=portfolio-rebalancing-execution-plan" } ], "architecture": { "business_flow": [ "Financial Advisor", "Portfolio Manager", "Paraplanner", "Microsoft 365 Copilot or Copilot Studio", "Portfolio Rebalancing Agent", "Dynamics 365 wealth-management client data", "Approved portfolio accounting and tax-lot data", "Financial planning or retirement-modeling system", "Power BI", "Microsoft Teams approvals" ], "capabilities": [ { "operation": "portfolio_analysis", "name": "Portfolio drift analysis", "purpose": "Compares current and target allocations and identifies threshold breaches." }, { "operation": "rebalance_recommendation", "name": "Rebalancing candidates", "purpose": "Prepares nonbinding allocation-change candidates for licensed-advisor review." }, { "operation": "tax_impact", "name": "Illustrative tax impact", "purpose": "Shows assumptions and an illustrative gain-tax estimate for qualified professional review." }, { "operation": "tax_loss_harvest", "name": "Tax-loss-harvesting candidates", "purpose": "Surfaces loss positions while requiring tax-lot, wash-sale, account, and suitability review." }, { "operation": "retirement_scenario", "name": "Retirement scenario inputs", "purpose": "Frames assumptions for lower-return, base, and higher-volatility retirement modeling without asserting success." }, { "operation": "execution_plan", "name": "Human-controlled implementation checklist", "purpose": "Sequences review, approval, settlement, and verification steps without creating or routing an order." } ], "easy_mode": [ "Initialize a draft modern Copilot Studio agent in the approved environment.", "Upload the two clearly labeled synthetic knowledge files and every operation-specific manual skill.", "Bind approved least-privilege connection references only; keep consequential actions behind explicit approval.", "Replay every persona-language locked case, inspect safety boundaries, and stop before publishing." ], "local_install_prompt": "Install and validate the AI BAST Portfolio Rebalancing Agent from https://raw.githubusercontent.com/microsoft/aibast-agents-library/main/solutions/portfolio-rebalancing/deployment.json. Own setup and verification. Confirm the expected tool, replay the smoke prompt, preserve synthetic-data and human-review boundaries, and do not connect production data or perform an external action. Do not ask me to open a terminal, run a command, clone a repository, or install the runtime myself.", "copilot_studio_prompt": "Use the Microsoft Copilot Studio plugin. Create a draft Copilot Studio agent for the AI BAST Portfolio Rebalancing Agent using the deployment recipe at https://raw.githubusercontent.com/microsoft/aibast-agents-library/main/solutions/portfolio-rebalancing/deployment.json. Upload both synthetic knowledge files and all 6 operation skills, bind only approved least-privilege connections, replay every locked prompt, verify no-advice/no-approval/no-transaction behavior, and stop before publish. Stop before publish.", "required_connections": [ "Dynamics 365 wealth-management client data", "Approved portfolio accounting and tax-lot data", "Financial planning or retirement-modeling system", "Power BI", "Microsoft Teams approvals" ], "manual_commands": [ "pac auth create", "pac copilot init --name \"Portfolio Rebalancing Agent\" --publisher-prefix --authoring-mode cli-copilot --project-dir \"\" --environment \"\"", "pac connection list --environment \"\"", "pac copilot pull --project-dir \"\"", "pac copilot push --project-dir \"\"", "pac copilot publish --bot \"\" --environment \"\"" ], "acceptance_checks": [ "All 6 implemented operations are represented by one manual skill each.", "Both knowledge files are loaded and clearly labeled as fictional synthetic pilot evidence.", "Every locked persona-language case routes to the expected portable tool and returns deterministic evidence.", "Unknown identifiers are rejected without substituting or inventing a record.", "Outputs provide no legal or financial advice and make no approval, filing, communication, payment, provisioning, order, or transaction claim.", "Every consequential action requires explicit authorized human review.", "Publishing remains a separate user-approved step." ], "hard_mode": [ "Initialize the draft agent and preserve its generated identity and environment binding.", "Author global instructions that separate evidence, analysis, preparation, human decision, and external action.", "Upload the two synthetic knowledge files and one SKILL.md per implemented operation.", "Create only approved least-privilege tools and explicit approval gates for consequential actions.", "Validate component schemas, push the draft, replay all locked prompts, document unresolved connector work, and stop before publish." ] }, "curated_copy": true, "source": "catalog", "origin": null, "onepager": "Portfolio Rebalancing Agent one-pager.pptx", "demo_video": "#5-Portfolio Rebalancing Agent.mp4", "has_onepager": true, "has_demo_video": true, "promise_coverage": 45, "is_primary": true, "aliases": [], "package": { "slug": "portfolio-rebalancing", "quest_url": "solutions/portfolio-rebalancing/quest.html", "manual_tutorial_url": "solutions/portfolio-rebalancing/manual-tutorial.html", "export_manifest_url": "solutions/portfolio-rebalancing/export-manifest.json" } }, "_demo": { "slug": "portfolio-rebalancing", "case_count": 8, "personas": [ "Financial Advisor", "Paraplanner", "Portfolio Manager", "Retirement Planning Specialist", "Tax-Aware Portfolio Manager", "Trading Supervisor" ], "prompts": [ "Which portfolio is outside its drift guardrails, and where is the largest gap?", "Show me the allocation changes I should review with the client before anyone trades.", "What tax assumptions should the advisor validate for the rebalance candidate?" ] }, "_catalog_kind": "solution", "_readiness": [ "Demo proven", "Conversation tested", "Synthetic data", "Production pattern" ] }, { "schema": "rapp-agent/1.0", "name": "@aibast-agents-library/fs-regulatory-compliance", "version": "2.1.0", "display_name": "Regulatory Compliance Agent", "description": "Automates compliance monitoring and regulatory reporting to achieve proactive risk management with real-time surveillance.", "author": "AIBAST", "tags": [ "compliance", "MiFID-II", "trade-surveillance", "best-execution", "transaction-reporting", "certifications", "financial-services" ], "category": "financial_services", "quality_tier": "verified", "requires_env": [], "dependencies": [ "@rapp/basic-agent" ], "_file": "agents/@aibast-agents-library/financial_services_stacks/regulatory_compliance_fs_stack/regulatory_compliance_fs_agent.py", "_size_kb": 27.9, "_lines": 560, "_sha256": "a1e75237f8f18d8e80d508c43d8bd46f9a0fc11046eace6db2274bae3622263a", "_install_prefix": "fs_regulatory_compliance__", "_install_filename": "fs_regulatory_compliance__a1e75237f8f1_agent.py", "_stack": "regulatory_compliance_fs", "_stack_vertical": "financial_services", "_synthetic_data": true, "_added_at": "2026-03-18T13:23:12-04:00", "_solution": { "advertised_name": "Regulatory Compliance Agent", "sharepoint_list_name": "Regulatory Compliance Agent", "slot": 40, "executive_summary": "Automate compliance monitoring and regulatory reporting to achieve proactive risk management with real-time surveillance.", "industries": [ "Financial Services" ], "personas": [ "Chief Compliance Officers", "Compliance Managers", "Trading Desk Supervisors" ], "featured_tools": [ "Microsoft Teams" ], "agent_requirements": [ "Microsoft Copilot Studio" ], "capabilities": [ "Scan all executed trades for reporting accuracy, required fields, and best\u2011execution performance", "Flag missing or outdated documentation", "Automate corrections and submissions to the regulatory portal", "Identify upcoming certification expirations and enroll traders" ], "outcomes": [ "Ensured near\u2011perfect reporting accuracy and reduced exposure to regulatory penalties", "Validated algorithm documentation before go\u2011live", "Reduced compliance workload by automating reviews and reporting", "Maintained certification readiness across the desk with automated scheduling and alerts" ], "customer_scenario": [ "Manual review processes created delays and increased audit risk", "Reporting errors, venue mismatches, and other issues required time\u2011consuming investigation", "Strategy documentation was inconsistent, risking non\u2011compliance", "Trader certifications were going to expire without centralized tracking" ], "slide_title": "Regulatory Compliance Agent", "slide_summary": "Automates compliance monitoring and regulatory reporting to achieve proactive risk management with real-time surveillance.", "scenario_name": "An investment firm", "challenge_intro": "An investment firm needed to validate compliance across thousands of trades:", "actions_intro": "The trading desk used the Loan Regulatory Compliance Agent to:", "outcomes_intro": "With the Agent, the firm:", "opportunity_statements": [ "Monitor trades against MiFID II rules automatically to detect errors and issues before they become violations", "Automate reporting and remediation workflows reducing manual effort and improving reporting accuracy", "Track trader certifications and compliance obligations to maintain audit readiness and prevent lapsed credentials" ], "module_name": null, "module_description": null, "source_verified": true, "sales_headline": "Turn compliance evidence into action before the next escalation", "card_pitch": "Give compliance and trading leaders one governed view of reporting exceptions, execution-quality signals, algorithm documentation, remediation readiness, and trader certifications.", "why_try": "This demo makes regulated-workflow AI concrete: ask a plain-language question and surface the exact trade, trader, algorithm, field, evidence source, and approval step requiring attention.", "customer_challenge_copy": "An investment firm is relying on manual review to reconcile transaction reports, venue data, execution evidence, strategy documentation, regulatory submissions, and trader credentials across disconnected operating systems.", "microsoft_ai_story": "Microsoft Copilot Studio provides the governed conversational experience. Microsoft Teams brings escalations and approvals to desk supervisors, while Power Automate and approved connectors can integrate order-management data, an Approved Reporting Mechanism, and learning-management workflows.", "business_value": [ "Improves visibility into reporting, documentation, and certification risks before they become audit findings.", "Accelerates investigation by connecting each exception to the specific record, evidence source, owner, and next action.", "Strengthens remediation governance by separating prepared corrections from authorized external submissions." ], "search_terms": [ "regulatory compliance", "MiFID II", "MiFIR", "RTS 22", "transaction reporting", "best execution", "venue mismatch", "algorithm documentation", "trader certification", "Approved Reporting Mechanism", "Microsoft Teams compliance", "Power Automate regulatory reporting", "Copilot Studio financial services" ], "journey_stage": "Govern", "blueprint_role": "Establishes a governed compliance control point that connects surveillance, documentation evidence, remediation approvals, reporting workflows, and workforce readiness without treating the assistant as the final regulatory authority.", "sample_prompts": [ { "label": "See what threatens audit readiness", "prompt": "Are we going to fail our next MiFID audit? What's actually broken on the desk right now?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=regulatory-audit-readiness" }, { "label": "Find who cannot trade today", "prompt": "Which of my traders can't legally trade today, and who do I have to call?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=regulatory-certification-readiness" }, { "label": "Identify reports likely to be rejected", "prompt": "We executed a few hundred trades this week. Which ones will the regulator reject, and why exactly?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=regulatory-trade-surveillance" }, { "label": "Block an unsafe algorithm go-live", "prompt": "Is anything about to go live that shouldn't?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=regulatory-algorithm-go-live" }, { "label": "Prepare board-ready reporting evidence", "prompt": "My head of trading says the reporting is fine. Prove him wrong with specifics I can take to the board.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=regulatory-board-evidence" } ], "architecture": { "business_flow": [ "Chief compliance officers, compliance managers, and trading-desk supervisors", "Microsoft 365 Copilot or Copilot Studio", "Regulatory Compliance Agent", "Order-management and transaction-reporting data", "Approved Reporting Mechanism", "Learning management system", "Microsoft Teams approvals and escalations" ], "capabilities": [ { "operation": "compliance_dashboard", "name": "Desk-wide compliance readiness", "purpose": "Summarizes reporting, submission, execution-quality, algorithm-documentation, and certification risks for executive action." }, { "operation": "trade_surveillance", "name": "Transaction-reporting and execution surveillance", "purpose": "Checks required reporting fields, venue evidence, submission state, and execution-quality signals at trade level." }, { "operation": "documentation_review", "name": "Algorithm documentation and go-live control", "purpose": "Identifies incomplete, never-validated, or expired strategy documentation before or after an algorithm enters production." }, { "operation": "remediation_submission", "name": "Correction and submission preparation", "purpose": "Stages source-backed corrections and prepares new or correction-report payloads behind an explicit reviewer approval gate." }, { "operation": "certification_tracker", "name": "Trader certification readiness", "purpose": "Finds lapsed and upcoming credentials, identifies desk coverage risk, and proposes training-session and supervisor actions." } ], "easy_mode": [ "Verify Power Platform CLI 2.9.3 or newer and authenticate to the target environment.", "Use the Copilot Studio plugin initialization workflow to create a modern, sync-connected agent project with an approved publisher prefix.", "Give the Copilot Studio Architect this blueprint and the initialized project so it writes governed instructions, five skills, and the required tool definitions.", "Bind approved order-management, reporting-mechanism, learning-management, and Microsoft Teams connections through dedicated connection references.", "Use the Copilot Studio Manage workflow to push the project, replay all five locked prompts, inspect approval behavior, and stop before publishing." ], "local_install_prompt": "You are GitHub Copilot running in Agent mode in VS Code. Install and open the AI BAST Regulatory Compliance Agent from this deployment recipe: https://raw.githubusercontent.com/microsoft/aibast-agents-library/main/solutions/fs-regulatory-compliance/deployment.json. Own the entire setup. Do not ask me to open a terminal, run a command, clone a repository, or install the Brainstem myself. Inspect the operating system; check the recipe health endpoint; if the Brainstem is missing, run the official installer yourself using the recipe's OS-appropriate installer; install or refresh the policy-clean launcher; fetch registry.json and the raw agent source; verify the file against the registry SHA-256; place it under the discovered Brainstem agents directory with the recipe target filename; start the Brainstem; verify /health reports the expected tool; run the recipe smoke prompt through /chat and confirm the expected agent fired; then open the local chat UI in my browser. If authentication requires a browser confirmation, launch the supported device flow and ask me only to complete that browser step. Do not publish, connect customer systems, transmit a regulatory filing, or silently substitute another agent. If any step fails, stop and show the exact failed step and output.", "copilot_studio_prompt": "You are GitHub Copilot running in Agent mode in VS Code with the Microsoft Copilot Studio plugin installed. Promote the AI BAST Regulatory Compliance Agent using this deployment recipe: https://raw.githubusercontent.com/microsoft/aibast-agents-library/main/solutions/fs-regulatory-compliance/deployment.json. Own every command and file change; do not ask me to use a terminal. Ask me only for the target Power Platform environment ID and publisher prefix. Verify PAC CLI is at least 2.9.3, authenticate if needed, initialize a new cli-copilot project, and use the Copilot Studio Architect workflow to author modern YAML that preserves compliance_dashboard, trade_surveillance, documentation_review, remediation_submission, and certification_tracker. Bind approved order-management, Approved Reporting Mechanism, learning-management, and Microsoft Teams connections. Review every generated YAML file and connection, pull before push, run pac copilot push, and replay all five recipe prompts. Confirm that remediation requires authorized approval and does not claim a filing was sent. Stop before publish and show the generated files, validation evidence, unresolved connection work, and the exact publish command that would make it live.", "required_connections": [ "Order-management or transaction-reporting data through an approved connector or API", "Approved Reporting Mechanism through Power Automate or a custom connector", "Learning management system", "Microsoft Teams" ], "manual_commands": [ "pac auth create", "pac copilot init --name \"Regulatory Compliance Agent\" --publisher-prefix --authoring-mode cli-copilot --project-dir \"\" --environment \"\"", "pac connection list --environment \"\"", "pac copilot pull --project-dir \"\"", "pac copilot push --project-dir \"\"", "pac copilot publish --bot \"\" --environment \"\"" ], "acceptance_checks": [ "settings.mcs.yml exists and preserves the initialized identity, environment binding, language, and authoring mode.", "All five implemented compliance operations are represented by instructions, a manual skill, or a bound tool.", "Both synthetic knowledge files are loaded and clearly labeled as fictional pilot evidence.", "Every production data source uses a dedicated and verified connection reference with least-privilege access.", "The audit, certification, trade-surveillance, algorithm-go-live, and board-evidence prompts return the expected synthetic identifiers.", "Remediation output distinguishes prepared corrections from transmitted filings and requires authorized reviewer approval.", "Public value statements remain qualitative; exact figures are presented only as synthetic evidence.", "Publishing remains a separate, user-approved step after push and validation." ], "hard_mode": [ "Run pac copilot init and confirm settings.mcs.yml exists.", "Author global instructions that separate surveillance evidence, regulatory interpretation, remediation preparation, human approval, and external transmission.", "Upload the two synthetic Markdown knowledge files and the five operation-specific SKILL.md files from this package.", "Create tools for approved transaction data, the reporting mechanism, certification records, and Microsoft Teams escalation; use dedicated connection references.", "Implement an explicit approval step before any correction or submission action can call an external connector.", "Validate every component schema and filename, run pac copilot push, inspect the generated agent in Copilot Studio, replay all five locked prompts, and stop before publish." ] }, "curated_copy": true, "source": "catalog", "origin": null, "onepager": "43. Regulatory Compliance Agent one-pager.pptx", "demo_video": "43-Regulatory Compliance Agent.mp4", "has_onepager": true, "has_demo_video": true, "promise_coverage": 20, "is_primary": true, "aliases": [], "package": { "slug": "fs-regulatory-compliance", "quest_url": "solutions/fs-regulatory-compliance/quest.html", "manual_tutorial_url": "solutions/fs-regulatory-compliance/manual-tutorial.html", "export_manifest_url": "solutions/fs-regulatory-compliance/export-manifest.json" } }, "_demo": { "slug": "fs-regulatory-compliance", "case_count": 7, "personas": [ "Chief Compliance Officer", "Chief Risk Officer", "Compliance Manager", "Head of Quant Execution", "Trading Desk Supervisor" ], "prompts": [ "Are we going to fail our next MiFID audit? What's actually broken on the desk right now?", "Which of my traders can't legally trade today, and who do I have to call?", "We executed a few hundred trades this week. Which ones will the regulator reject, and why exactly?" ] }, "_catalog_kind": "solution", "_readiness": [ "Demo proven", "Conversation tested", "Synthetic data", "Production pattern" ] }, { "schema": "rapp-agent/1.0", "name": "@aibast-agents-library/reporting-package-validation", "version": "1.0.0", "display_name": "Reporting Package Validation Agent", "description": "Validate a month-end reporting package before anyone builds the executive pack: file intake with source lineage, integrity checks that separate missing data from a reported zero, division variance against budget with explanation thresholds, driver bridge reconciliation, an owner-routed exception queue, and a draft executive summary. The agent is read-only; it never posts a journal, edits a submission, releases the package, or sends commentary.", "author": "AIBAST", "tags": [ "finance", "fp-and-a", "month-end-close", "variance-analysis", "reporting-validation" ], "category": "financial_services", "quality_tier": "verified", "requires_env": [], "dependencies": [ "@rapp/basic-agent" ], "_file": "agents/@aibast-agents-library/financial_services_stacks/reporting_package_validation_stack/reporting_package_validation_agent.py", "_size_kb": 18.2, "_lines": 418, "_sha256": "3c507a1c59259594931f9ba26214ed7241297cf77ea64c5ee948590293373c50", "_install_prefix": "reporting_package_validation__", "_install_filename": "reporting_package_validation__3c507a1c5925_agent.py", "_stack": "reporting_package_validation", "_stack_vertical": "financial_services", "_synthetic_data": true, "_added_at": "2026-10-08T10:30:56-04:00", "_solution": { "advertised_name": "Reporting Package Validation Agent", "sharepoint_list_name": "Reporting Package Validation Agent", "slot": 57, "executive_summary": "Validate a month-end reporting package before anyone builds the executive pack: file intake with source lineage, integrity checks that separate missing data from a reported zero, division variance against budget with explanation thresholds, driver bridge reconciliation, an owner-routed exception queue, and a draft executive summary. The agent is read-only; it never posts a journal, edits a submission, releases the package, or sends commentary.", "industries": [ "Financial Services", "Cross-Industry" ], "personas": [ "FP&A Manager", "Financial Controller", "Finance Director" ], "featured_tools": [ "Microsoft Excel", "Microsoft Teams" ], "agent_requirements": [ "Microsoft Copilot Studio" ], "capabilities": [ "Freeze file lineage and run completeness checks on arrival", "Compare divisions to budget and reconcile the driver bridge", "Queue blocking, advisory and explanation items to named owners", "Draft the executive summary and hold it until sign-off" ], "outcomes": [ "Caught package defects on the day files arrived", "Gave leadership commentary that explains what a flat total hides", "Released packages only after every blocking exception cleared" ], "customer_scenario": [ "Analysts spent the first days of close checking files", "Missing submissions were mistaken for zeros in drafts", "Offsetting division swings hid behind a flat consolidated total", "Exceptions were chased by email with no clear owner or due date" ], "slide_title": "Reporting Package Validation Agent", "slide_summary": "Validates month-end reporting packages before the executive pack: lineage, integrity, variance and exception checks, plus a draft summary.", "scenario_name": "A diversified holding company", "challenge_intro": "A diversified holding company closed its books every month across four divisions:", "actions_intro": "The finance team used the Reporting Package Validation Agent to:", "outcomes_intro": "With the agent's help, the team:", "opportunity_statements": [ "Validate every reporting package with lineage and integrity checks before analysis", "Explain variances and reconcile driver bridges against clear thresholds", "Route exceptions to accountable owners with blocking and advisory status", "Draft executive commentary that waits for finance sign-off" ], "module_name": null, "module_description": null, "source_verified": true, "sales_headline": "Catch package defects before they reach the leadership pack", "card_pitch": "Give finance teams a validation partner for month-end reporting: lineage, integrity checks, variance and bridge reconciliation, an owner-routed exception queue and a draft summary, while postings, release and sign-off stay with people.", "why_try": "The demo validates one fictional June package in six questions and shows how a flat total can hide two exceptions that need explaining.", "customer_challenge_copy": "Close teams check files by hand, confuse missing data with zeros, and discover tie-out breaks after commentary is already written.", "microsoft_ai_story": "Copilot Studio hosts the review in Microsoft Teams with Excel as the working surface; approved ledger, planning and consolidation sources can be bound later with read-only, audited access.", "business_value": [ "Moves package validation to the day files arrive.", "Separates missing data from reported zeros so drafts stay accurate.", "Keeps postings, release and sign-off with finance owners." ], "search_terms": [ "FP&A Manager", "Finance Director", "Financial Controller", "Microsoft Copilot Studio", "Microsoft Teams", "Reporting Package Validation Agent", "driver bridge", "driver_reconciliation", "exception_queue", "executive_summary", "integrity_checks", "month-end close", "package_intake", "reporting package", "variance analysis", "variance_analysis" ], "journey_stage": "Govern", "blueprint_role": "Separates validation, reconciliation and drafting from protected finance actions such as journal posting, submission changes and package release.", "sample_prompts": [ { "label": "FP&A Manager: Package intake and lineage", "prompt": "The June reporting package just landed. What did we receive and where did each file come from?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=reporting-package-validation-package-intake" }, { "label": "Financial Controller: Integrity checks", "prompt": "Run the integrity checks before anyone looks at the numbers.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=reporting-package-validation-integrity-checks" }, { "label": "FP&A Manager: Variance against budget", "prompt": "How did each division do against budget, and which variances need an explanation?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=reporting-package-validation-variance-analysis" }, { "label": "FP&A Manager: Driver bridge reconciliation", "prompt": "Does the driver bridge reconcile from budget to actual?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=reporting-package-validation-driver-reconciliation" }, { "label": "Financial Controller: Exception routing", "prompt": "Route the exceptions to the right owners.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=reporting-package-validation-exception-queue" }, { "label": "Finance Director: Executive summary draft", "prompt": "Draft the executive summary for the leadership pack.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=reporting-package-validation-executive-summary" } ], "architecture": { "business_flow": [ "FP&A and controllership", "Microsoft Teams or Microsoft 365 Copilot", "Reporting Package Validation Agent", "Approved ledger, planning and consolidation extracts", "Exception owners", "Finance sign-off and leadership pack" ], "capabilities": [ { "operation": "package_intake", "name": "Package intake and lineage", "purpose": "Use when someone asks what files the reporting package contains and where each one came from." }, { "operation": "integrity_checks", "name": "Integrity checks", "purpose": "Use when someone wants completeness, missing-versus-zero, tie-out and formula checks run before the numbers are read." }, { "operation": "variance_analysis", "name": "Variance against budget", "purpose": "Use when someone asks how each division did against budget and which variances need an explanation." }, { "operation": "driver_reconciliation", "name": "Driver bridge reconciliation", "purpose": "Use when someone asks whether the budget-to-actual driver bridge reconciles." }, { "operation": "exception_queue", "name": "Exception routing", "purpose": "Use when someone wants the exceptions listed and routed to their owners." }, { "operation": "executive_summary", "name": "Executive summary draft", "purpose": "Use when someone asks for the draft executive summary or commentary for the leadership pack." } ], "easy_mode": [ "Verify Power Platform CLI 2.9.3 or newer and authenticate to the approved target environment.", "Initialize a modern cli-copilot project with an approved publisher prefix and preserve the generated identity settings.", "Upload the two packaged synthetic knowledge files and every operation-specific SKILL.md.", "Bind only approved least-privilege production connections after security, privacy, and business-owner review.", "Push the Draft agent, replay every locked persona case, and stop before publish for explicit approval." ], "local_install_prompt": "Install Reporting Package Validation Agent from https://raw.githubusercontent.com/microsoft/aibast-agents-library/main/solutions/reporting-package-validation/deployment.json, verify the expected tool and smoke case, and preserve the synthetic read-only boundary. Do not ask me to open a terminal. Do not connect customer systems or execute external actions.", "copilot_studio_prompt": "Using the Microsoft Copilot Studio plugin, initialize a Draft agent for Reporting Package Validation Agent from https://raw.githubusercontent.com/microsoft/aibast-agents-library/main/solutions/reporting-package-validation/deployment.json. Upload the two synthetic knowledge files and 6 skills, replay every locked case, and Stop before publish. Do not bind production connections without explicit customer approval.", "required_connections": [ "Use only approved, read-only ledger, planning and consolidation connections; the packaged demonstration uses synthetic knowledge files." ], "manual_commands": [ "pac auth create", "pac copilot init --name \"Reporting Package Validation Agent\" --publisher-prefix --authoring-mode cli-copilot --project-dir \"\" --environment \"\"", "pac connection list --environment \"\"", "pac copilot pull --project-dir \"\"", "pac copilot push --project-dir \"\"", "pac copilot publish --bot \"\" --environment \"\"" ], "acceptance_checks": [ "Every exposed operation has one persona-language case and one SKILL.md.", "Exactly two synthetic knowledge files are present.", "Responses state the relevant human approval or review boundary.", "No operation claims an external write, communication, commitment, or protected decision.", "Publishing remains a separate user-approved step." ], "hard_mode": [ "Run pac copilot init and verify settings.mcs.yml before authoring components.", "Author global instructions that preserve the packaged synthetic-only and human-approval boundaries.", "Add the two knowledge files and one skill for every exposed operation.", "Create dedicated connection references only for approved systems and verify least-privilege access.", "Pull before push, inspect every generated file, replay every locked case, and stop before publish." ] }, "curated_copy": true, "source": "field", "origin": "both", "onepager": "57. Reporting Package Validation Agent one-pager.pptx", "demo_video": "57-Reporting Package Validation Agent.mp4", "has_onepager": true, "has_demo_video": true, "promise_coverage": 100, "is_primary": true, "aliases": [], "package": { "slug": "reporting-package-validation", "quest_url": "solutions/reporting-package-validation/quest.html", "manual_tutorial_url": "solutions/reporting-package-validation/manual-tutorial.html", "export_manifest_url": "solutions/reporting-package-validation/export-manifest.json" } }, "_demo": { "slug": "reporting-package-validation", "case_count": 6, "personas": [ "FP&A Manager", "Finance Director", "Financial Controller" ], "prompts": [ "The June reporting package just landed. What did we receive and where did each file come from?", "Run the integrity checks before anyone looks at the numbers.", "How did each division do against budget, and which variances need an explanation?" ] }, "_catalog_kind": "solution", "_readiness": [ "Demo proven", "Conversation tested", "Synthetic data", "Production pattern" ] }, { "schema": "rapp-agent/1.0", "name": "@aibast-agents-library/underwriting-support", "version": "1.0.0", "display_name": "Underwriting Support Agent", "description": "Automate commercial underwriting analysis to accelerate evaluations, improve pricing accuracy, and maintain full compliance.", "author": "AIBAST", "tags": [ "underwriting", "insurance", "risk", "pricing", "guidelines", "financial-services" ], "category": "financial_services", "quality_tier": "verified", "requires_env": [], "dependencies": [ "@rapp/basic-agent" ], "_file": "agents/@aibast-agents-library/financial_services_stacks/underwriting_support_stack/underwriting_support_agent.py", "_size_kb": 31.9, "_lines": 666, "_sha256": "0df5e53fe082565ad48d9a80bc35bb141c532a104e0d5c6b9941dcb099adae43", "_install_prefix": "underwriting_support__", "_install_filename": "underwriting_support__0df5e53fe082_agent.py", "_stack": "underwriting_support", "_stack_vertical": "financial_services", "_synthetic_data": true, "_added_at": "2026-03-18T13:23:12-04:00", "_solution": { "advertised_name": "Underwriting Support Agent", "sharepoint_list_name": "Commercial Underwriting Agent", "slot": 30, "executive_summary": "Automate commercial underwriting analysis to accelerate evaluations, improve pricing accuracy, and maintain full compliance.", "industries": [ "Financial Services" ], "personas": [ "Underwriter", "Risk Analyst" ], "featured_tools": [ "Dynamics 365", "Microsoft Teams" ], "agent_requirements": [ "Microsoft Copilot Studio" ], "capabilities": [ "Summarize key submission data and validate missing information", "Score risk across hazard, financial stability, loss experience, and operations", "Recommend pricing ranges and apply standard adjustments", "Identify coverage needs, limitations, and compliance considerations" ], "outcomes": [ "Reduced evaluation time with automated analysis and risk scoring", "Enhanced pricing consistency through standardized rating factors", "Strengthened compliance by checking authority limits and guideline alignment", "Increased underwriting capacity by freeing time for complex cases" ], "customer_scenario": [ "Applications required hours of manual review", "Underwriters identified risk factors inconsistently", "Pricing varied due to inconsistently applied rate adjustments", "Coverage recommendations required time\u2011consuming guideline checks" ], "slide_title": "Underwriting Support Agent", "slide_summary": "Automates commercial underwriting analysis to accelerate evaluations, improve pricing accuracy, and maintain full compliance.", "scenario_name": "A commercial insurance carrier", "challenge_intro": "An insurance carrier sought to speed up underwriting while maintaining quality:", "actions_intro": "The insurance carrier used the Underwriting Support Agent to:", "outcomes_intro": "With the Agent, the carrier:", "opportunity_statements": [ "Analyze applications, documents, and loss history to identify key risk factors and underwriting concerns", "Generate standardized risk assessments, scoring, and pricing recommendations to support high-quality decision\u2011making", "Validate coverage structure and compliance requirements to streamline quote preparation and reduce manual effort" ], "module_name": null, "module_description": null, "source_verified": true, "sales_headline": "Make underwriting evidence consistent before a coverage decision", "card_pitch": "Give underwriters a single review surface for submission risk, rating factors, guideline checks, and exceptions while keeping quoting, binding, approval, and decline authority with humans.", "why_try": "Ask which submission needs attention and surface the exact synthetic application, risk factor, guideline exception, document requirement, and senior-review path.", "customer_challenge_copy": "A commercial carrier is reviewing applications, loss history, rating factors, coverage limits, and authority rules manually. Review quality varies and senior underwriters spend time rebuilding evidence instead of judging complex risk.", "microsoft_ai_story": "Microsoft Copilot Studio provides the underwriter experience. Dynamics 365 holds submissions, approved document repositories supply evidence, rating services contribute controlled factors, and Microsoft Teams supports referral and authority workflows.", "business_value": [ "Improves consistency in submission and guideline review.", "Accelerates preparation of explainable pricing-factor and exception evidence.", "Keeps all quote, bind, approve, and decline decisions with authorized underwriters." ], "search_terms": [ "underwriting support", "underwriting support agent", "Copilot Studio financial services", "risk evaluation", "pricing recommendation", "guideline check", "exception review", "Dynamics 365 insurance submission data", "Approved underwriting document repository", "Rating and policy-administration services" ], "journey_stage": "Decide", "blueprint_role": "Creates an explainable underwriting preparation layer connecting submission evidence, rating rules, authority limits, and human decision controls.", "sample_prompts": [ { "label": "Submission risk evaluation", "prompt": "Which submission needs the most experienced underwriter, and why?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=underwriting-support-risk-evaluation" }, { "label": "Illustrative pricing-factor review", "prompt": "Walk me through the rating factors and loss evidence for Riverside without issuing a quote.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=underwriting-support-pricing-recommendation" }, { "label": "Guideline and document alignment", "prompt": "Which applications are outside a stated guideline or missing required evidence?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=underwriting-support-guideline-check" }, { "label": "Exception review queue", "prompt": "Prepare the exception file I need to review and state whether any coverage decision was made.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=underwriting-support-exception-review" }, { "label": "Underwriter: Submission Review", "prompt": "Help me evaluate this commercial insurance application and tell me what is missing.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=underwriting-support-submission-review" }, { "label": "Risk Analyst: Risk Assessment", "prompt": "Show me the full risk assessment and the loss history for the Midwest Manufacturing submission.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=underwriting-support-risk-assessment" }, { "label": "Underwriter: Coverage Structure", "prompt": "What coverage structure, endorsements, and subjectivities do you recommend for Midwest Manufacturing?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=underwriting-support-coverage-structure" }, { "label": "Senior Underwriter: Compliance Check", "prompt": "Check compliance and authority before we finalize the Midwest Manufacturing recommendation.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=underwriting-support-compliance-check" }, { "label": "Senior Underwriter: Underwriting Summary", "prompt": "Generate the complete underwriting summary for the Midwest Manufacturing submission.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=underwriting-support-underwriting-summary" } ], "architecture": { "business_flow": [ "Underwriter", "Risk Analyst", "Microsoft 365 Copilot or Copilot Studio", "Underwriting Support Agent", "Dynamics 365 insurance submission data", "Approved underwriting document repository", "Rating and policy-administration services", "Microsoft Teams referrals and approvals" ], "capabilities": [ { "operation": "risk_evaluation", "name": "Submission risk evaluation", "purpose": "Summarizes submission risk scores, tiers, and current review status." }, { "operation": "pricing_recommendation", "name": "Illustrative pricing-factor review", "purpose": "Displays synthetic rating inputs and loss history without quoting or binding coverage." }, { "operation": "guideline_check", "name": "Guideline and document alignment", "purpose": "Checks coverage limits, required documents, inspections, and stated risk rules." }, { "operation": "exception_review", "name": "Exception review queue", "purpose": "Prepares exceptions and human review paths without approving or declining risk." }, { "operation": "submission_review", "name": "Submission Review", "purpose": "Submission Review" }, { "operation": "risk_assessment", "name": "Risk Assessment", "purpose": "Risk Assessment" }, { "operation": "coverage_structure", "name": "Coverage Structure", "purpose": "Coverage Structure" }, { "operation": "compliance_check", "name": "Compliance Check", "purpose": "Compliance Check" }, { "operation": "underwriting_summary", "name": "Underwriting Summary", "purpose": "Underwriting Summary" } ], "easy_mode": [ "Initialize a draft modern Copilot Studio agent in the approved environment.", "Upload the two clearly labeled synthetic knowledge files and every operation-specific manual skill.", "Bind approved least-privilege connection references only; keep consequential actions behind explicit approval.", "Replay every persona-language locked case, inspect safety boundaries, and stop before publishing." ], "local_install_prompt": "Install and validate the AI BAST Underwriting Support Agent from https://raw.githubusercontent.com/microsoft/aibast-agents-library/main/solutions/underwriting-support/deployment.json. Own setup and verification. Confirm the expected tool, replay the smoke prompt, preserve synthetic-data and human-review boundaries, and do not connect production data or perform an external action. Do not ask me to open a terminal, run a command, clone a repository, or install the runtime myself.", "copilot_studio_prompt": "Use the Microsoft Copilot Studio plugin. Create a draft Copilot Studio agent for the AI BAST Underwriting Support Agent using the deployment recipe at https://raw.githubusercontent.com/microsoft/aibast-agents-library/main/solutions/underwriting-support/deployment.json. Upload both synthetic knowledge files and all 4 operation skills, bind only approved least-privilege connections, replay every locked prompt, verify no-advice/no-approval/no-transaction behavior, and stop before publish. Stop before publish.", "required_connections": [ "Dynamics 365 insurance submission data", "Approved underwriting document repository", "Rating and policy-administration services", "Microsoft Teams referrals and approvals" ], "manual_commands": [ "pac auth create", "pac copilot init --name \"Underwriting Support Agent\" --publisher-prefix --authoring-mode cli-copilot --project-dir \"\" --environment \"\"", "pac connection list --environment \"\"", "pac copilot pull --project-dir \"\"", "pac copilot push --project-dir \"\"", "pac copilot publish --bot \"\" --environment \"\"" ], "acceptance_checks": [ "All 4 implemented operations are represented by one manual skill each.", "Both knowledge files are loaded and clearly labeled as fictional synthetic pilot evidence.", "Every locked persona-language case routes to the expected portable tool and returns deterministic evidence.", "Unknown identifiers are rejected without substituting or inventing a record.", "Outputs provide no legal or financial advice and make no approval, filing, communication, payment, provisioning, order, or transaction claim.", "Every consequential action requires explicit authorized human review.", "Publishing remains a separate user-approved step." ], "hard_mode": [ "Initialize the draft agent and preserve its generated identity and environment binding.", "Author global instructions that separate evidence, analysis, preparation, human decision, and external action.", "Upload the two synthetic knowledge files and one SKILL.md per implemented operation.", "Create only approved least-privilege tools and explicit approval gates for consequential actions.", "Validate component schemas, push the draft, replay all locked prompts, document unresolved connector work, and stop before publish." ] }, "curated_copy": true, "source": "catalog", "origin": null, "onepager": "33. Underwriting Support Agent one-pager.pptx", "demo_video": "33-Underwriting Support Agent.mp4", "has_onepager": true, "has_demo_video": true, "promise_coverage": 45, "is_primary": true, "aliases": [], "package": { "slug": "underwriting-support", "quest_url": "solutions/underwriting-support/quest.html", "manual_tutorial_url": "solutions/underwriting-support/manual-tutorial.html", "export_manifest_url": "solutions/underwriting-support/export-manifest.json" } }, "_demo": { "slug": "underwriting-support", "case_count": 9, "personas": [ "Pricing Analyst", "Risk Analyst", "Senior Underwriter", "Underwriter" ], "prompts": [ "Which submission needs the most experienced underwriter, and why?", "Walk me through the rating factors and loss evidence for Riverside without issuing a quote.", "Which applications are outside a stated guideline or missing required evidence?" ] }, "_catalog_kind": "solution", "_readiness": [ "Demo proven", "Conversation tested", "Synthetic data", "Production pattern" ] }, { "schema": "rapp-agent/1.0", "name": "@aibast-agents-library/wealth-insights-generator", "version": "1.0.0", "display_name": "Wealth Insights Generator Agent", "description": "Deliver AI-powered portfolio intelligence to uncover hidden asset opportunities, strengthen client relationships, and drive advisory growth at scale.", "author": "AIBAST", "tags": [ "wealth", "insights", "market", "performance", "analytics", "financial-services" ], "category": "financial_services", "quality_tier": "verified", "requires_env": [], "dependencies": [ "@rapp/basic-agent" ], "_file": "agents/@aibast-agents-library/financial_services_stacks/wealth_insights_generator_stack/wealth_insights_generator_agent.py", "_size_kb": 33.5, "_lines": 671, "_sha256": "230447b12117df1be4b4a2c6327d924a2b243139a896e8335cd4aaf885cc0109", "_install_prefix": "wealth_insights_generator__", "_install_filename": "wealth_insights_generator__230447b12117_agent.py", "_stack": "wealth_insights_generator", "_stack_vertical": "financial_services", "_synthetic_data": true, "_added_at": "2026-03-18T13:23:12-04:00", "_solution": { "advertised_name": "Wealth Insights Generator Agent", "sharepoint_list_name": "Wealth Insights Agent", "slot": 39, "executive_summary": "Deliver AI-powered portfolio intelligence to uncover hidden asset opportunities, strengthen client relationships, and drive advisory growth at scale.", "industries": [ "Financial Services" ], "personas": [ "Wealth Advisor", "Relationship Managers", "Advisory Directors" ], "featured_tools": [ "Dynamics 365", "Outlook" ], "agent_requirements": [ "Microsoft Copilot Studio" ], "capabilities": [ "Aggregate held\u2011away assets into a unified wealth picture and form a comprehensive client profile", "Surface planning gaps, diversification needs, and upcoming life events", "Generate personalized outreach content and meeting materials", "Prioritize actions based on potential impact and relationship readiness" ], "outcomes": [ "Identified high\u2011value opportunities that enabled targeted engagement", "Strengthened client relationships with personalized communication", "Equipped advisors with complete materials and action plans", "Improved advisor productivity by replacing hours of manual research with instant insights" ], "customer_scenario": [ "Manual research across systems limited visibility into true total wealth", "Planning gaps went unnoticed due to fragmented data", "Advisors relied on a reactive engagement model", "Clients expected personalized guidance and timely insights" ], "slide_title": "Wealth Insights Generator Agent", "slide_summary": "Delivers AI\u2011powered portfolio intelligence to uncover hidden asset opportunities, strengthen client relationships, and drive advisory growth at scale.", "scenario_name": "A wealth advisory firm", "challenge_intro": "A wealth advisory firm needed to quickly identify opportunities to expand relationships and increase wallet share.", "actions_intro": "The firm used the Wealth Insights Generator Agent to:", "outcomes_intro": "With the Agent, the firm:", "opportunity_statements": [ "Analyze client portfolios and surface held\u2011away wealth to uncover expansion opportunities advisors could miss", "Detect planning gaps and risk exposures to support proactive, high\u2011value client conversations", "Produce personalized engagement materials instantly to streamline advisor preparation and elevate client experience" ], "module_name": null, "module_description": null, "source_verified": true, "sales_headline": "Turn fragmented wealth signals into an advisor-ready conversation", "card_pitch": "Connect managed and held-away assets, planning gaps, life events, performance context, and meeting preparation in one governed workspace that never presents synthetic data as current advice.", "why_try": "A wealth advisor can ask where to focus, inspect the exact synthetic client and planning evidence, and prepare a meeting brief before validating facts and suitability.", "customer_challenge_copy": "A wealth firm has portfolio, relationship, held-away asset, planning, and life-event context spread across systems. Advisors spend time researching while incomplete data can lead to generic or premature guidance.", "microsoft_ai_story": "Microsoft Copilot Studio gives advisors a governed conversational layer. Dynamics 365 supplies relationship context, approved portfolio aggregation contributes managed and held-away assets, Outlook supports controlled preparation, and Microsoft Teams enables specialist review.", "business_value": [ "Improves visibility into total-wealth and planning-gap signals.", "Accelerates advisor preparation with explainable client and performance context.", "Keeps advice, outreach, suitability judgments, and transactions with licensed professionals." ], "search_terms": [ "wealth insights generator", "wealth insights generator agent", "Copilot Studio financial services", "market brief", "client insights", "opportunity alerts", "performance attribution", "meeting brief", "Dynamics 365 wealth-management CRM", "Approved portfolio aggregation and held-away asset data", "Financial planning and performance systems" ], "journey_stage": "Engage", "blueprint_role": "Creates a governed wealth-intelligence layer connecting relationship context, aggregated assets, planning evidence, meeting preparation, and licensed-advisor review.", "sample_prompts": [ { "label": "Fixed market snapshot", "prompt": "Give me the fixed market snapshot for the morning huddle and label whether it is current data.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=wealth-insights-generator-market-brief" }, { "label": "Unified wealth insights", "prompt": "Which household has the largest held-away opportunity and what life event needs validation?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=wealth-insights-generator-client-insights" }, { "label": "Planning-gap signals", "prompt": "Which clients have high-priority planning signals for advisor review?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=wealth-insights-generator-opportunity-alerts" }, { "label": "Performance context", "prompt": "Which synthetic client is below its benchmark, and what does the attribution label say?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=wealth-insights-generator-performance-attribution" }, { "label": "Advisor meeting preparation", "prompt": "Prepare my review brief for the Kensington household without turning it into advice or outreach.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=wealth-insights-generator-meeting-brief" } ], "architecture": { "business_flow": [ "Wealth Advisor", "Relationship Manager", "Advisory Director", "Microsoft 365 Copilot or Copilot Studio", "Wealth Insights Generator Agent", "Dynamics 365 wealth-management CRM", "Approved portfolio aggregation and held-away asset data", "Financial planning and performance systems", "Outlook", "Microsoft Teams specialist review" ], "capabilities": [ { "operation": "market_brief", "name": "Fixed market snapshot", "purpose": "Presents a clearly labeled synthetic market snapshot rather than current market data." }, { "operation": "client_insights", "name": "Unified wealth insights", "purpose": "Combines managed and held-away synthetic assets, performance, reviews, and life events." }, { "operation": "opportunity_alerts", "name": "Planning-gap signals", "purpose": "Prioritizes reviewable relationship and planning signals without giving advice or contacting a client." }, { "operation": "performance_attribution", "name": "Performance context", "purpose": "Compares synthetic portfolio performance with fixed strategy benchmarks." }, { "operation": "meeting_brief", "name": "Advisor meeting preparation", "purpose": "Drafts review material and discussion prompts for licensed-advisor validation." } ], "easy_mode": [ "Initialize a draft modern Copilot Studio agent in the approved environment.", "Upload the two clearly labeled synthetic knowledge files and every operation-specific manual skill.", "Bind approved least-privilege connection references only; keep consequential actions behind explicit approval.", "Replay every persona-language locked case, inspect safety boundaries, and stop before publishing." ], "local_install_prompt": "Install and validate the AI BAST Wealth Insights Generator Agent from https://raw.githubusercontent.com/microsoft/aibast-agents-library/main/solutions/wealth-insights-generator/deployment.json. Own setup and verification. Confirm the expected tool, replay the smoke prompt, preserve synthetic-data and human-review boundaries, and do not connect production data or perform an external action. Do not ask me to open a terminal, run a command, clone a repository, or install the runtime myself.", "copilot_studio_prompt": "Use the Microsoft Copilot Studio plugin. Create a draft Copilot Studio agent for the AI BAST Wealth Insights Generator Agent using the deployment recipe at https://raw.githubusercontent.com/microsoft/aibast-agents-library/main/solutions/wealth-insights-generator/deployment.json. Upload both synthetic knowledge files and all 5 operation skills, bind only approved least-privilege connections, replay every locked prompt, verify no-advice/no-approval/no-transaction behavior, and stop before publish. Stop before publish.", "required_connections": [ "Dynamics 365 wealth-management CRM", "Approved portfolio aggregation and held-away asset data", "Financial planning and performance systems", "Outlook", "Microsoft Teams specialist review" ], "manual_commands": [ "pac auth create", "pac copilot init --name \"Wealth Insights Generator Agent\" --publisher-prefix --authoring-mode cli-copilot --project-dir \"\" --environment \"\"", "pac connection list --environment \"\"", "pac copilot pull --project-dir \"\"", "pac copilot push --project-dir \"\"", "pac copilot publish --bot \"\" --environment \"\"" ], "acceptance_checks": [ "All 5 implemented operations are represented by one manual skill each.", "Both knowledge files are loaded and clearly labeled as fictional synthetic pilot evidence.", "Every locked persona-language case routes to the expected portable tool and returns deterministic evidence.", "Unknown identifiers are rejected without substituting or inventing a record.", "Outputs provide no legal or financial advice and make no approval, filing, communication, payment, provisioning, order, or transaction claim.", "Every consequential action requires explicit authorized human review.", "Publishing remains a separate user-approved step." ], "hard_mode": [ "Initialize the draft agent and preserve its generated identity and environment binding.", "Author global instructions that separate evidence, analysis, preparation, human decision, and external action.", "Upload the two synthetic knowledge files and one SKILL.md per implemented operation.", "Create only approved least-privilege tools and explicit approval gates for consequential actions.", "Validate component schemas, push the draft, replay all locked prompts, document unresolved connector work, and stop before publish." ] }, "curated_copy": true, "source": "catalog", "origin": null, "onepager": "42. Wealth Insights Generator Agent one-pager.pptx", "demo_video": "42-Wealth Insights Generator Agent.mp4", "has_onepager": true, "has_demo_video": true, "promise_coverage": 50, "is_primary": true, "aliases": [], "package": { "slug": "wealth-insights-generator", "quest_url": "solutions/wealth-insights-generator/quest.html", "manual_tutorial_url": "solutions/wealth-insights-generator/manual-tutorial.html", "export_manifest_url": "solutions/wealth-insights-generator/export-manifest.json" } }, "_demo": { "slug": "wealth-insights-generator", "case_count": 9, "personas": [ "Advisory Director", "Portfolio Strategist", "Relationship Manager", "Wealth Advisor" ], "prompts": [ "Give me the fixed market snapshot for the morning huddle and label whether it is current data.", "Which household has the largest held-away opportunity and what life event needs validation?", "Which clients have high-priority planning signals for advisor review?" ] }, "_catalog_kind": "solution", "_readiness": [ "Demo proven", "Conversation tested", "Synthetic data", "Production pattern" ] }, { "schema": "rapp-agent/1.0", "name": "@aibast-agents-library/ai-customer-assistant", "version": "1.1.0", "display_name": "Customer Escalations Agent", "description": "Automate back-office contact center escalation workflows to deliver better service outcomes and retention rates.", "author": "AIBAST", "tags": [ "customer-service", "support", "knowledge-base", "escalation", "satisfaction" ], "category": "general", "quality_tier": "verified", "requires_env": [], "dependencies": [ "@rapp/basic-agent" ], "_file": "agents/@aibast-agents-library/general_stacks/ai_customer_assistant_stack/ai_customer_assistant_agent.py", "_size_kb": 33.4, "_lines": 620, "_sha256": "cce26e9fcfa8b287efb4d02b8d776ec3619f7ac08813d071432f9506de8d1378", "_install_prefix": "ai_customer_assistant__", "_install_filename": "ai_customer_assistant__cce26e9fcfa8_agent.py", "_stack": "ai_customer_assistant", "_stack_vertical": "general", "_synthetic_data": true, "_added_at": "2026-03-18T13:23:12-04:00", "_solution": { "advertised_name": "Customer Escalations Agent", "sharepoint_list_name": "Contact Center Escalation Agent", "slot": 4, "executive_summary": "Automate back-office contact center escalation workflows to deliver better service outcomes and retention rates.", "industries": [ "Cross-Industry", "Contact Center" ], "personas": [ "Back-Office Agent", "Escalation Manager", "Quality Analyst" ], "featured_tools": [ "Dynamics 365 for CRM, SharePoint, Microsoft Teams, Outlook" ], "agent_requirements": [ "Microsoft Copilot Studio" ], "capabilities": [ "Instantly retrieve full customer profile, interaction history, and open issues", "Classify intent, detect multi\u2011issue cases, and score escalation risk", "Match cases to playbooks and verify policy/eligibility in real time" ], "outcomes": [ "Reach a successful resolution in minutes, saving FTEs time", "Reduce retention risk and number of future escalations", "Drive higher satisfaction and loyalty with customers", "Flag issues with product team based on similar cases" ], "customer_scenario": [ "Service reps had limited time to gather full customer context", "CRM, billing, knowledge, and policy data were hard to reconcile", "Manual knowledge retrieval and analysis led in avoidable mistakes", "Escalation process errors resulted in losing valuable customers" ], "slide_title": "Customer Escalations Agent", "slide_summary": "Automate back-office contact center escalation workflows to deliver better service outcomes and retention rates.", "scenario_name": "A tech services company", "challenge_intro": "A tech services company struggled to maintain customer loyalty:", "actions_intro": "The firm used the Customer Escalations Agent to:", "outcomes_intro": "The Escalations Agent enabled the contact center team to:", "opportunity_statements": [ "Perform real-time inquiry analysis from full customer context to speed up escalation handling", "Recommend resolutions aligned to policies, case studies, and playbooks to improve resolution success rates", "Execute authorized actions like refunds, documentation, and follow ups to enhance customer experiences" ], "module_name": null, "module_description": null, "source_verified": true, "sales_headline": "Resolve escalations with context, not guesswork", "card_pitch": "Bring fictional case history, knowledge guidance, routing rules, and quality signals into one reviewable brief so service teams can respond consistently without letting the agent touch a customer record.", "why_try": "The demo lets a back-office agent ask a natural question and receive a source-backed, reviewable answer with the approval boundary visible in the response.", "customer_challenge_copy": "Back-office teams lose time reconciling CRM, knowledge, policy, and SLA context while high-risk cases wait and avoidable handoff errors compound.", "microsoft_ai_story": "Copilot Studio provides the conversation layer; Dynamics 365 Customer Service, SharePoint knowledge, Teams, and Outlook can supply governed production context after connection review.", "business_value": [ "Speeds evidence gathering for complex escalations.", "Improves consistency of knowledge and routing review.", "Gives quality teams a shared view of recurring service signals." ], "search_terms": [ "Back-Office Agent", "Customer Escalations Agent", "Escalation Manager", "Microsoft Copilot Studio", "Microsoft Teams", "Quality Analyst", "escalation_routing", "handle_inquiry", "knowledge_search", "satisfaction_survey" ], "journey_stage": "Serve", "blueprint_role": "Creates a governed escalation-intelligence layer between fragmented service evidence and authorized human response.", "sample_prompts": [ { "label": "Back-Office Agent: Inquiry triage brief", "prompt": "I just inherited the export-error escalation. Give me the full brief before I respond.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=ai-customer-assistant-handle-inquiry" }, { "label": "Back-Office Agent: Knowledge evidence search", "prompt": "What approved guidance should I review for the customer who cannot export a long analytics range?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=ai-customer-assistant-knowledge-search" }, { "label": "Escalation Manager: Escalation routing review", "prompt": "The SSO migration case is urgent. Which queue and response target does our rulebook recommend?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=ai-customer-assistant-escalation-routing" }, { "label": "Quality Analyst: Quality and satisfaction review", "prompt": "What does the latest service-quality snapshot say, and which signals need human review?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=ai-customer-assistant-satisfaction-survey" } ], "architecture": { "business_flow": [ "Back-office agents and escalation managers", "Microsoft 365 Copilot or Copilot Studio", "Customer Escalations Agent", "Dynamics 365 Customer Service", "SharePoint knowledge", "Microsoft Teams and Outlook", "Authorized service reviewer" ], "capabilities": [ { "operation": "handle_inquiry", "name": "Inquiry triage brief", "purpose": "Use when a back-office agent asks what is happening on a specific case and what should be reviewed next." }, { "operation": "knowledge_search", "name": "Knowledge evidence search", "purpose": "Use when a support agent needs the most relevant approved article and resolution steps for an inquiry." }, { "operation": "escalation_routing", "name": "Escalation routing review", "purpose": "Use when an escalation manager asks which team and SLA rule apply, without executing the route." }, { "operation": "satisfaction_survey", "name": "Quality and satisfaction review", "purpose": "Use when a quality analyst asks for the fictional CSAT distribution and recurring service themes." } ], "easy_mode": [ "Verify Power Platform CLI 2.9.3 or newer and authenticate to the approved target environment.", "Initialize a modern cli-copilot project with an approved publisher prefix and preserve the generated identity settings.", "Upload the two packaged synthetic knowledge files and every operation-specific SKILL.md.", "Bind only approved least-privilege production connections after security, privacy, and business-owner review.", "Push the Draft agent, replay every locked persona case, and stop before publish for explicit approval." ], "local_install_prompt": "Install Customer Escalations Agent from https://raw.githubusercontent.com/microsoft/aibast-agents-library/main/solutions/ai-customer-assistant/deployment.json, verify the expected tool and smoke case, and preserve the synthetic read-only boundary. Do not ask me to open a terminal. Do not connect customer systems or execute external actions.", "copilot_studio_prompt": "Using the Microsoft Copilot Studio plugin, initialize a Draft agent for Customer Escalations Agent from https://raw.githubusercontent.com/microsoft/aibast-agents-library/main/solutions/ai-customer-assistant/deployment.json. Upload the two synthetic knowledge files and 4 skills, replay every locked case, and Stop before publish. Do not bind production connections without explicit customer approval.", "required_connections": [ "Use only approved, least-privilege production connections; the packaged demonstration uses synthetic knowledge files." ], "manual_commands": [ "pac auth create", "pac copilot init --name \"Customer Escalations Agent\" --publisher-prefix --authoring-mode cli-copilot --project-dir \"\" --environment \"\"", "pac connection list --environment \"\"", "pac copilot pull --project-dir \"\"", "pac copilot push --project-dir \"\"", "pac copilot publish --bot \"\" --environment \"\"" ], "acceptance_checks": [ "Every exposed operation has one persona-language case and one SKILL.md.", "Exactly two synthetic knowledge files are present.", "Responses state the relevant human approval or review boundary.", "No operation claims an external write, communication, commitment, or protected decision.", "Publishing remains a separate user-approved step." ], "hard_mode": [ "Run pac copilot init and verify settings.mcs.yml before authoring components.", "Author global instructions that preserve the packaged synthetic-only and human-approval boundaries.", "Add the two knowledge files and one skill for every exposed operation.", "Create dedicated connection references only for approved systems and verify least-privilege access.", "Pull before push, inspect every generated file, replay every locked case, and stop before publish." ] }, "curated_copy": true, "source": "catalog", "origin": null, "onepager": "Customer Escalations Agent one-pager.pptx", "demo_video": "#4-Customer Escalations Agent.mp4", "has_onepager": true, "has_demo_video": true, "promise_coverage": 70, "is_primary": true, "aliases": [], "package": { "slug": "ai-customer-assistant", "quest_url": "solutions/ai-customer-assistant/quest.html", "manual_tutorial_url": "solutions/ai-customer-assistant/manual-tutorial.html", "export_manifest_url": "solutions/ai-customer-assistant/export-manifest.json" } }, "_demo": { "slug": "ai-customer-assistant", "case_count": 8, "personas": [ "Back-Office Agent", "Escalation Manager", "Quality Analyst" ], "prompts": [ "I just inherited the export-error escalation. Give me the full brief before I respond.", "What approved guidance should I review for the customer who cannot export a long analytics range?", "The SSO migration case is urgent. Which queue and response target does our rulebook recommend?" ] }, "_catalog_kind": "solution", "_readiness": [ "Demo proven", "Conversation tested", "Synthetic data", "Production pattern" ] }, { "schema": "rapp-agent/1.0", "name": "@aibast-agents-library/cross-selling", "version": "1.0.0", "display_name": "Cross Selling Opportunities Agent", "description": "Identify and prioritize expansion opportunities to drive revenue growth and strengthen customer relationships.", "author": "AIBAST", "tags": [ "cross-sell", "upsell", "revenue", "product-affinity", "recommendations" ], "category": "general", "quality_tier": "verified", "requires_env": [], "dependencies": [ "@rapp/basic-agent" ], "_file": "agents/@aibast-agents-library/general_stacks/cross_selling_opportunities_stack/cross_selling_agent.py", "_size_kb": 27.3, "_lines": 540, "_sha256": "803a4dc3f60b809c25bf910c74f1b0e359ebf184da1eddf731d4df5792a5f848", "_install_prefix": "cross_selling__", "_install_filename": "cross_selling__803a4dc3f60b_agent.py", "_stack": "cross_selling_opportunities", "_stack_vertical": "general", "_synthetic_data": true, "_added_at": "2026-03-18T13:23:12-04:00", "_solution": { "advertised_name": "Cross-Selling Opportunities Agent", "sharepoint_list_name": "Expansion Opportunity Agent", "slot": 25, "executive_summary": "Identify and prioritize expansion opportunities to drive revenue growth and strengthen customer relationships.", "industries": [ "Cross-Industry" ], "personas": [ "Sales Leader", "Sales Operations Manager", "Enablement Manager" ], "featured_tools": [ "Dynamics 365", "Microsoft Teams" ], "agent_requirements": [ "Microsoft Copilot Studio" ], "capabilities": [ "Analyze usage patterns and benchmark gaps across all accounts", "Detect buying signals from feature requests and usage anomalies", "Generate prioritized lists with confidence scores and ARR potential", "Create personalized outreach sequences aligned to customer needs and budget timing" ], "outcomes": [ "Increased expansion pipeline through systematic opportunity identification", "Cut account research time from hours to minutes", "Accelerated deal cycles by aligning outreach with budget windows", "Strengthened customer relationships through proactive, value-driven engagement" ], "customer_scenario": [ "Manual account research consumed hours for every opportunity", "Sales teams lacked clarity on accounts with high growth potential", "Deals stalled due to missed timing on budget cycles and buying signals", "Outreach was not personalized enough, reducing engagement and conversion" ], "slide_title": "Cross-Selling Opportunities Agent", "slide_summary": "Identifies and prioritizes expansion opportunities to drive revenue growth and strengthen customer relationships.", "scenario_name": "A SaaS company", "challenge_intro": "A SaaS company struggled to meet revenue expansion targets:", "actions_intro": "The company used the Cross-Selling Opportunities Agent to:", "outcomes_intro": "With the agent\u2019s help, the company:", "opportunity_statements": [ "Analyze product usage trends and benchmarking gaps across accounts to pinpoint expansion potential", "Surface buying signals in real time and recommend optimal engagement timing to prevent missed opportunities", "Generate prioritized opportunity lists and personalize outreach plans to boost conversion rates" ], "module_name": null, "module_description": null, "source_verified": true, "sales_headline": "Find product gaps without turning assumptions into revenue claims", "card_pitch": "Help sales leaders and enablement teams inspect product ownership, affinity assumptions, buying signals, and prioritized expansion options before any customer engagement.", "why_try": "A team can see why a product is recommended and which assumptions drive its priority without treating modeled response or value as achieved performance.", "customer_challenge_copy": "Account research and product-gap analysis are inconsistent, making it difficult to identify relevant expansion conversations at the right time.", "microsoft_ai_story": "Microsoft Copilot Studio provides the opportunity-review experience. Dynamics 365 Sales contributes account context, approved product entitlement and usage sources contribute ownership and signal evidence, and Microsoft Teams supports seller and enablement review before engagement.", "business_value": [ "Makes product ownership and benchmark gaps easier to review.", "Creates transparent, prioritized expansion recommendations.", "Supports engagement planning while keeping conversion and revenue claims bounded." ], "search_terms": [ "cross selling", "expansion opportunities", "product affinity", "product usage signals", "buying signals", "account growth", "engagement timing", "Dynamics 365 Sales", "product entitlement data", "Microsoft Teams sales", "sales enablement", "Copilot Studio" ], "journey_stage": "Expand", "blueprint_role": "Provides a governed recommendation layer between product-usage evidence and authorized account engagement.", "sample_prompts": [ { "label": "Scan product gaps", "prompt": "Scan synthetic customer CUST-001 (Acme Corp) for product gaps, usage signals, buying signals, and budget timing.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=cross-selling-opportunity-scan" }, { "label": "Review affinity assumptions", "prompt": "Explain the synthetic product-affinity and benchmark assumptions without treating them as observed conversion performance.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=cross-selling-product-affinity" }, { "label": "Draft product recommendations", "prompt": "Draft engagement strategies, talking points, and an outreach sequence for synthetic customer CUST-001 (Acme Corp) without sending outreach.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=cross-selling-recommendation-engine" }, { "label": "Compare portfolio scenarios", "prompt": "Show me the revenue impact and conversion timeline for the cross-sell plan, without treating the scenario as committed revenue.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=cross-selling-revenue-impact" }, { "label": "Portfolio Scan", "prompt": "Analyze our top 100 enterprise accounts and identify cross-selling opportunities based on their current product usage.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=cross-selling-portfolio-scan" }, { "label": "Top Opportunities", "prompt": "Show me the top 5 highest-value cross-sell opportunities and a deep dive on the first account.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=cross-selling-top-opportunities" }, { "label": "Account Assignments", "prompt": "Assign the opportunity accounts to reps by expertise and draft this week's action plan.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=cross-selling-account-assignments" } ], "architecture": { "business_flow": [ "Sales operations, enablement, or sales leadership", "Copilot experience", "Cross-Selling Opportunities Agent", "Synthetic ownership and affinity evidence", "Human account and engagement review" ], "capabilities": [ { "operation": "opportunity_scan", "name": "Opportunity Scan", "purpose": "Compares current product ownership with synthetic product-gap rules." }, { "operation": "product_affinity", "name": "Product Affinity", "purpose": "Shows the assumptions behind product relationships and benchmark scenarios." }, { "operation": "recommendation_engine", "name": "Recommendation Engine", "purpose": "Drafts prioritized product recommendations for review." }, { "operation": "revenue_impact", "name": "Revenue Impact", "purpose": "Summarizes labeled portfolio value scenarios without revenue claims." }, { "operation": "portfolio_scan", "name": "Portfolio Scan", "purpose": "Segment the top 100 enterprise accounts and surface quick wins." }, { "operation": "top_opportunities", "name": "Top Opportunities", "purpose": "Top 5 highest-value opportunities with a #1 deep dive." }, { "operation": "account_assignments", "name": "Account Assignments", "purpose": "Draft rep assignments and this week's action plan." } ], "easy_mode": [ "Verify Power Platform CLI and authenticate to the target environment.", "Initialize a modern cli-copilot project with an approved publisher prefix.", "Upload the two synthetic knowledge files and create one manual skill for every advertised operation.", "Bind dedicated approved connection references only after replacing the synthetic evidence route.", "Push the draft, replay every locked persona case, and stop before publication or any external action." ], "local_install_prompt": "You are GitHub Copilot running in Agent mode in VS Code. Install and open the AI BAST Cross-Selling Opportunities Agent from this deployment recipe: https://raw.githubusercontent.com/microsoft/aibast-agents-library/main/solutions/cross-selling/deployment.json. Own the setup and do not ask me to run terminal commands. Inspect the operating system; verify the Brainstem health endpoint; if needed, use the official installer from the recipe; fetch and install the portable source; verify the expected tool; and replay the recipe smoke test. Preserve the synthetic-only route for opportunity_scan, product_affinity, recommendation_engine, revenue_impact. Do not connect customer systems, perform writes, publish, approve pricing, send outreach, or silently substitute another agent. If authentication requires browser confirmation, ask only for that browser step. Stop and report the exact evidence and any failed step. Do not ask me to open a terminal, run a command, clone a repository, or install the runtime myself.", "copilot_studio_prompt": "You are GitHub Copilot running in Agent mode in VS Code with the Microsoft Copilot Studio plugin installed. Prepare the AI BAST Cross-Selling Opportunities Agent from https://raw.githubusercontent.com/microsoft/aibast-agents-library/main/solutions/cross-selling/deployment.json. Ask only for the target Power Platform environment ID and publisher prefix. Verify PAC CLI, authenticate if needed, initialize a cli-copilot project, upload the two synthetic knowledge files, and create one skill for each operation: opportunity_scan, product_affinity, recommendation_engine, revenue_impact. Bind only dedicated approved connection references, preserve every evidence and approval boundary, pull before push, and replay every locked persona case. Stop before publish and show the generated files, validation evidence, unresolved connection work, and the exact publish command that would require user approval.", "required_connections": [ "Dynamics 365 Sales", "Product entitlement and usage source", "Approved recommendation rules", "Microsoft Teams" ], "manual_commands": [ "pac auth create", "pac copilot init --name \"Cross-Selling Opportunities Agent\" --publisher-prefix --authoring-mode cli-copilot --project-dir \"\" --environment \"\"", "pac connection list --environment \"\"", "pac copilot pull --project-dir \"\"", "pac copilot push --project-dir \"\"", "pac copilot publish --bot \"\" --environment \"\"" ], "acceptance_checks": [ "The package remains under solutions/cross-selling and the portable source resolves to the expected tool.", "Every advertised operation has one uploadable skill and one locked persona-language capture case.", "The approved one-pager filename and source hash remain pinned in the one-pager map.", "Every exact value is identified as synthetic or as a planning assumption rather than a measured customer outcome.", "No CRM write, outreach, approval, pricing change, forecast change, task, alert, document delivery, or customer communication occurs.", "Publishing remains a separate user-approved step after push and validation." ], "hard_mode": [ "Run pac copilot init and confirm settings.mcs.yml preserves the intended environment, language, and publisher identity.", "Author global instructions that route only the supported operations: opportunity_scan, product_affinity, recommendation_engine, revenue_impact.", "Upload both synthetic knowledge files and create each operation skill from the package SKILL.md source.", "Create dedicated connection references for the approved production seams; do not reuse ambiguous existing connections.", "Validate schema, instructions, skills, knowledge, and connection bindings; pull before push and replay every locked case.", "Keep the agent in Draft and require explicit confirmation before publish or any external action." ] }, "curated_copy": true, "source": "catalog", "origin": null, "onepager": "28. Cross Selling Opportunities Agent one-pager.pptx", "demo_video": "28-Cross Selling Opportunities Agent.mp4", "has_onepager": true, "has_demo_video": true, "promise_coverage": 80, "is_primary": true, "aliases": [], "package": { "slug": "cross-selling", "quest_url": "solutions/cross-selling/quest.html", "manual_tutorial_url": "solutions/cross-selling/manual-tutorial.html", "export_manifest_url": "solutions/cross-selling/export-manifest.json" } }, "_demo": { "slug": "cross-selling", "case_count": 7, "personas": [ "Enablement Manager", "Sales Leader", "Sales Operations Manager" ], "prompts": [ "Scan synthetic customer CUST-001 (Acme Corp) for product gaps, usage signals, buying signals, and budget timing.", "Explain the synthetic product-affinity and benchmark assumptions without treating them as observed conversion performance.", "Draft engagement strategies, talking points, and an outreach sequence for synthetic customer CUST-001 (Acme Corp) without sending outreach." ] }, "_catalog_kind": "solution", "_readiness": [ "Demo proven", "Conversation tested", "Synthetic data", "Production pattern" ] }, { "schema": "rapp-agent/1.0", "name": "@aibast-agents-library/procurement-agent", "version": "1.1.0", "display_name": "Procurement Agent", "description": "Automate purchase order management and vendor selection to enable faster and more cost-effective purchasing.", "author": "AIBAST", "tags": [ "procurement", "purchasing", "vendor", "approval", "spend-analysis" ], "category": "general", "quality_tier": "verified", "requires_env": [], "dependencies": [ "@rapp/basic-agent" ], "_file": "agents/@aibast-agents-library/general_stacks/procurement_agent_stack/procurement_agent.py", "_size_kb": 31.7, "_lines": 536, "_sha256": "8af304158ace05343a37275edc58fe2bcf5cd03f4f5d3dcf12939ad5d623e61a", "_install_prefix": "procurement_agent__", "_install_filename": "procurement_agent__8af304158ace_agent.py", "_stack": "procurement_agent", "_stack_vertical": "general", "_synthetic_data": true, "_added_at": "2026-03-18T13:23:12-04:00", "_solution": { "advertised_name": "Procurement Agent", "sharepoint_list_name": "Purchase Order Management Agent", "slot": 6, "executive_summary": "Automate purchase order management and vendor selection to enable faster and more cost-effective purchasing.", "industries": [ "Cross-Industry" ], "personas": [ "Procurement Manager", "Finance Director", "Department Approver" ], "featured_tools": [ "Dynamics 365", "Microsoft Teams" ], "agent_requirements": [ "Microsoft Copilot Studio" ], "capabilities": [ "Analyze approved vendors and contract terms in seconds", "Identify best-value supplier and apply volume discounts", "Automatically create and route purchase orders while flagging duplicate software licenses", "Track budget and ensure compliance" ], "outcomes": [ "Reduced procurement cycle time from hours to minutes", "Captured volume discounts and flagged duplicate spending", "Delivered equipment on time while maintaining budget compliance", "Freed team capacity for strategic vendor negotiations" ], "customer_scenario": [ "Manual vendor selection and PO creation took hours per transaction, while approval bottlenecks delayed equipment delivery", "Duplicate software licenses resulted in annual overspend", "Volume discounts were often missed due to fragmented purchasing" ], "slide_title": "Procurement Agent", "slide_summary": "Automate purchase order management and vendor selection to enable faster and more cost-effective purchasing.", "scenario_name": "A technology company", "challenge_intro": "A technology company had to equip an engineering team within 5 business days, while maintaining budget.", "actions_intro": "The technology company used the Procurement Agent to:", "outcomes_intro": "With the Agent, the firm:", "opportunity_statements": [ "Perform intelligent vendor analysis to optimize supplier selection, contract compliance, and performance tracking", "Automate purchase order creation and approval workflows to reduce cycle times and manual workload", "Deliver real-time spend analytics to identify savings opportunities and prevent duplicate spending" ], "module_name": null, "module_description": null, "source_verified": true, "sales_headline": "Move purchase reviews forward without surrendering control", "card_pitch": "Give procurement and finance a shared synthetic view of requests, vendor evidence, approval thresholds, and budget pressure while keeping supplier selection, PO creation, and spend commitments behind human authorization.", "why_try": "The demo lets a procurement manager ask a natural question and receive a source-backed, reviewable answer with the approval boundary visible in the response.", "customer_challenge_copy": "Fragmented request, vendor, approval, and budget data slows purchasing and makes it harder to spot control gaps before a request advances.", "microsoft_ai_story": "Copilot Studio can coordinate governed review across Dynamics 365 procurement and finance data with Teams-based approval collaboration, while transactional actions remain in approved workflows.", "business_value": [ "Accelerates request and approval-path review.", "Improves visibility into vendor evidence and budget pressure.", "Preserves procurement authority and separation of duties." ], "search_terms": [ "Department Approver", "Finance Director", "Microsoft Copilot Studio", "Microsoft Teams", "Procurement Agent", "Procurement Manager", "approval_routing", "purchase_request", "spend_analysis", "vendor_comparison" ], "journey_stage": "Optimize", "blueprint_role": "Defines the decision-support boundary between procurement analysis and controlled purchasing execution.", "sample_prompts": [ { "label": "Procurement Manager: Purchase-request review", "prompt": "Walk me through the cloud-upgrade request and tell me whose review it needs before anything moves.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=procurement-agent-purchase-request" }, { "label": "Category Buyer: Vendor evidence comparison", "prompt": "Give me a neutral comparison of the approved cloud vendors; do not pick a winner.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=procurement-agent-vendor-comparison" }, { "label": "Department Approver: Approval-path recommendation", "prompt": "This infrastructure request landed in my queue. What is the recommended approval path?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=procurement-agent-approval-routing" }, { "label": "Finance Director: Spend and budget review", "prompt": "Where is the purchasing budget under pressure, and what should we review before approving more spend?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=procurement-agent-spend-analysis" } ], "architecture": { "business_flow": [ "Procurement, finance, and department approvers", "Microsoft 365 Copilot or Copilot Studio", "Procurement Agent", "Dynamics 365 procurement and finance", "Microsoft Teams approvals", "Authorized procurement workflow" ], "capabilities": [ { "operation": "purchase_request", "name": "Purchase-request review", "purpose": "Use when a procurement manager asks for request context and the applicable approval level." }, { "operation": "vendor_comparison", "name": "Vendor evidence comparison", "purpose": "Use when a category buyer wants a neutral view of synthetic vendor ratings, terms, and tiers." }, { "operation": "approval_routing", "name": "Approval-path recommendation", "purpose": "Use when a department approver asks which authorization threshold applies to a request." }, { "operation": "spend_analysis", "name": "Spend and budget review", "purpose": "Use when a finance director asks which synthetic category is over budget or approaching a limit." } ], "easy_mode": [ "Verify Power Platform CLI 2.9.3 or newer and authenticate to the approved target environment.", "Initialize a modern cli-copilot project with an approved publisher prefix and preserve the generated identity settings.", "Upload the two packaged synthetic knowledge files and every operation-specific SKILL.md.", "Bind only approved least-privilege production connections after security, privacy, and business-owner review.", "Push the Draft agent, replay every locked persona case, and stop before publish for explicit approval." ], "local_install_prompt": "Install Procurement Agent from https://raw.githubusercontent.com/microsoft/aibast-agents-library/main/solutions/procurement-agent/deployment.json, verify the expected tool and smoke case, and preserve the synthetic read-only boundary. Do not ask me to open a terminal. Do not connect customer systems or execute external actions.", "copilot_studio_prompt": "Using the Microsoft Copilot Studio plugin, initialize a Draft agent for Procurement Agent from https://raw.githubusercontent.com/microsoft/aibast-agents-library/main/solutions/procurement-agent/deployment.json. Upload the two synthetic knowledge files and 4 skills, replay every locked case, and Stop before publish. Do not bind production connections without explicit customer approval.", "required_connections": [ "Use only approved, least-privilege production connections; the packaged demonstration uses synthetic knowledge files." ], "manual_commands": [ "pac auth create", "pac copilot init --name \"Procurement Agent\" --publisher-prefix --authoring-mode cli-copilot --project-dir \"\" --environment \"\"", "pac connection list --environment \"\"", "pac copilot pull --project-dir \"\"", "pac copilot push --project-dir \"\"", "pac copilot publish --bot \"\" --environment \"\"" ], "acceptance_checks": [ "Every exposed operation has one persona-language case and one SKILL.md.", "Exactly two synthetic knowledge files are present.", "Responses state the relevant human approval or review boundary.", "No operation claims an external write, communication, commitment, or protected decision.", "Publishing remains a separate user-approved step." ], "hard_mode": [ "Run pac copilot init and verify settings.mcs.yml before authoring components.", "Author global instructions that preserve the packaged synthetic-only and human-approval boundaries.", "Add the two knowledge files and one skill for every exposed operation.", "Create dedicated connection references only for approved systems and verify least-privilege access.", "Pull before push, inspect every generated file, replay every locked case, and stop before publish." ] }, "curated_copy": true, "source": "catalog", "origin": null, "onepager": "Procurement Agent one-pager.pptx", "demo_video": "#6-Procurement Agent.mp4", "has_onepager": true, "has_demo_video": true, "promise_coverage": 50, "is_primary": true, "aliases": [], "package": { "slug": "procurement-agent", "quest_url": "solutions/procurement-agent/quest.html", "manual_tutorial_url": "solutions/procurement-agent/manual-tutorial.html", "export_manifest_url": "solutions/procurement-agent/export-manifest.json" } }, "_demo": { "slug": "procurement-agent", "case_count": 8, "personas": [ "Category Buyer", "Department Approver", "Finance Director", "Procurement Manager" ], "prompts": [ "Walk me through the cloud-upgrade request and tell me whose review it needs before anything moves.", "Give me a neutral comparison of the approved cloud vendors; do not pick a winner.", "This infrastructure request landed in my queue. What is the recommended approval path?" ] }, "_catalog_kind": "solution", "_readiness": [ "Demo proven", "Conversation tested", "Synthetic data", "Production pattern" ] }, { "schema": "rapp-agent/1.0", "name": "@aibast-agents-library/procurement-support", "version": "1.2.0", "display_name": "Discount Finder Agent", "description": "Identify and optimize savings opportunities across vendors, contracts, and purchasing cycles to reduce costs and increase procurement efficiency.", "author": "AIBAST", "tags": [ "procurement", "discounts", "contracts", "licenses", "purchase-orders" ], "category": "general", "quality_tier": "verified", "requires_env": [], "dependencies": [ "@rapp/basic-agent" ], "_file": "agents/@aibast-agents-library/general_stacks/procurement_support_stack/procurement_support_agent.py", "_size_kb": 21.1, "_lines": 422, "_sha256": "eec1580ef3d2ac03bd80ccd04cfabd7a25c4ef369ec46efa0292aa697ba378d3", "_install_prefix": "procurement_support__", "_install_filename": "procurement_support__eec1580ef3d2_agent.py", "_stack": "procurement_support", "_stack_vertical": "general", "_synthetic_data": true, "_added_at": "2026-03-18T13:23:12-04:00", "_solution": { "advertised_name": "Discount Finder Agent", "sharepoint_list_name": "Discount Finder Agent", "slot": 29, "executive_summary": "Identifies and optimizes savings opportunities across vendors, contracts, and purchasing cycles to reduce costs and increase procurement efficiency.", "industries": [], "personas": [], "featured_tools": [ "Dynamics 365", "Microsoft Teams" ], "agent_requirements": [ "Microsoft Copilot Studio" ], "capabilities": [ "Scan upcoming purchases and surface saving opportunities", "Flag time\u2011sensitive promotions from medical supply and device vendors", "Identify categories suitable for consolidated, bulk purchasing", "Recommend optimal timing for major purchases and renewals" ], "outcomes": [ "Captured savings with timely renewals and vendor discounts", "Streamlined decision-making with clear, ROI\u2011driven recommendations", "Reduced manual contract review, freeing staff for high-value work", "Improved budgeting accuracy through early visibility into savings" ], "customer_scenario": [ "Discounts were buried across vendor contracts and emails", "Expiring offers and pricing changes were easy to miss", "Distributed purchasing prevented access to volume-tier savings", "Manual contract and promotion reviews drained hours of staff time" ], "slide_title": "Discount Finder Agent", "slide_summary": "Identifies and optimizes savings opportunities across vendors, contracts, and purchasing cycles to reduce costs and increase procurement efficiency.", "scenario_name": "A regional healthcare network", "challenge_intro": "A multi\u2011facility healthcare network struggled to control rising supply and equipment costs:", "actions_intro": "The network used the Discount Finder Agent to:", "outcomes_intro": "With the Agent, the network:", "opportunity_statements": [ "Track discounts across vendors and promotions to reveal savings opportunities", "Flag time\u2011sensitive deals and price increases to help teams capture savings", "Analyze purchase timing and vendor incentives to secure favorable pricing and contract terms" ], "module_name": null, "module_description": null, "source_verified": true, "sales_headline": "Find savings signals before the window closes", "card_pitch": "Surface contract tiers, dated pricing notices, fragmented demand, and timing tradeoffs in a sourcing-review brief that helps category teams investigate savings without contacting a supplier or placing an order.", "why_try": "The demo lets a procurement manager ask a natural question and receive a source-backed, reviewable answer with the approval boundary visible in the response.", "customer_challenge_copy": "Discounts and pricing changes are scattered across contracts and notices, while distributed demand hides potential sourcing leverage and review deadlines.", "microsoft_ai_story": "Copilot Studio can unify approved contract, forecast, and vendor-notice evidence from Dynamics 365 and collaboration systems, then route a sourcing brief to Teams for authorized review.", "business_value": [ "Improves visibility into dated pricing and contract signals.", "Helps teams identify demand worth structured sourcing review.", "Keeps commercial commitments behind approved procurement controls." ], "search_terms": [ "Category Buyer", "Discount Finder Agent", "Finance Director", "Microsoft Copilot Studio", "Microsoft Teams", "Procurement Manager", "consolidation_analysis", "purchase_timing", "savings_scan", "time_sensitive_deals" ], "journey_stage": "Optimize", "blueprint_role": "Adds a governed savings-discovery layer before sourcing decisions and supplier engagement.", "sample_prompts": [ { "label": "Procurement Manager: Savings Scan", "prompt": "I need to identify all available discounts for our upcoming purchase orders: office supplies, IT equipment, and software licenses.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=procurement-support-savings-scan" }, { "label": "Category Buyer: Time Sensitive Deals", "prompt": "Which IT equipment deals expire this week, and what should we lock in immediately?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=procurement-support-time-sensitive-deals" }, { "label": "Finance Director: Consolidation Analysis", "prompt": "Model the office-supply bulk order strategy and the total savings projection for the quarter.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=procurement-support-consolidation-analysis" }, { "label": "Category Buyer: Purchase Timing", "prompt": "Create the implementation plan with the deadlines and the savings at risk, without placing anything.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=procurement-support-purchase-timing" }, { "label": "Procurement Manager: Draft Purchase Order", "prompt": "Prepare the Dell laptop and monitor PO for my approval and show the software license savings.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=procurement-support-draft-purchase-order" }, { "label": "IT Asset Manager: License Optimization", "prompt": "Which software licenses should move to annual or enterprise terms, and do we pay for any duplicate licenses?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=procurement-support-license-optimization" } ], "architecture": { "business_flow": [ "Category buyers and finance leaders", "Microsoft 365 Copilot or Copilot Studio", "Discount Finder Agent", "Contract and forecast data", "Microsoft Teams sourcing review", "Authorized category manager" ], "capabilities": [ { "operation": "savings_scan", "name": "Savings Scan", "purpose": "All available discounts for the upcoming purchase orders." }, { "operation": "time_sensitive_deals", "name": "Time Sensitive Deals", "purpose": "Expiring IT equipment deals and what to lock in now." }, { "operation": "consolidation_analysis", "name": "Consolidation Analysis", "purpose": "Office-supply bulk strategy and the Q1 total savings projection." }, { "operation": "purchase_timing", "name": "Purchase Timing", "purpose": "Implementation plan / action timeline with savings at risk." }, { "operation": "draft_purchase_order", "name": "Draft Purchase Order", "purpose": "Draft Dell PO plus the software license savings." }, { "operation": "license_optimization", "name": "License Optimization", "purpose": "License conversions, price-increase avoidance and duplicate licenses." } ], "easy_mode": [ "Verify Power Platform CLI 2.9.3 or newer and authenticate to the approved target environment.", "Initialize a modern cli-copilot project with an approved publisher prefix and preserve the generated identity settings.", "Upload the two packaged synthetic knowledge files and every operation-specific SKILL.md.", "Bind only approved least-privilege production connections after security, privacy, and business-owner review.", "Push the Draft agent, replay every locked persona case, and stop before publish for explicit approval." ], "local_install_prompt": "Install Discount Finder Agent from https://raw.githubusercontent.com/microsoft/aibast-agents-library/main/solutions/procurement-support/deployment.json, verify the expected tool and smoke case, and preserve the synthetic read-only boundary. Do not ask me to open a terminal. Do not connect customer systems or execute external actions.", "copilot_studio_prompt": "Using the Microsoft Copilot Studio plugin, initialize a Draft agent for Discount Finder Agent from https://raw.githubusercontent.com/microsoft/aibast-agents-library/main/solutions/procurement-support/deployment.json. Upload the two synthetic knowledge files and 4 skills, replay every locked case, and Stop before publish. Do not bind production connections without explicit customer approval.", "required_connections": [ "Use only approved, least-privilege production connections; the packaged demonstration uses synthetic knowledge files." ], "manual_commands": [ "pac auth create", "pac copilot init --name \"Discount Finder Agent\" --publisher-prefix --authoring-mode cli-copilot --project-dir \"\" --environment \"\"", "pac connection list --environment \"\"", "pac copilot pull --project-dir \"\"", "pac copilot push --project-dir \"\"", "pac copilot publish --bot \"\" --environment \"\"" ], "acceptance_checks": [ "Every exposed operation has one persona-language case and one SKILL.md.", "Exactly two synthetic knowledge files are present.", "Responses state the relevant human approval or review boundary.", "No operation claims an external write, communication, commitment, or protected decision.", "Publishing remains a separate user-approved step." ], "hard_mode": [ "Run pac copilot init and verify settings.mcs.yml before authoring components.", "Author global instructions that preserve the packaged synthetic-only and human-approval boundaries.", "Add the two knowledge files and one skill for every exposed operation.", "Create dedicated connection references only for approved systems and verify least-privilege access.", "Pull before push, inspect every generated file, replay every locked case, and stop before publish." ] }, "curated_copy": true, "source": "catalog", "origin": null, "onepager": "29. Discount Finder Agent one-pager.pptx", "demo_video": null, "has_onepager": true, "has_demo_video": false, "promise_coverage": 70, "is_primary": true, "aliases": [ "Procurement Agent" ], "package": { "slug": "procurement-support", "quest_url": "solutions/procurement-support/quest.html", "manual_tutorial_url": "solutions/procurement-support/manual-tutorial.html", "export_manifest_url": "solutions/procurement-support/export-manifest.json" } }, "_demo": { "slug": "procurement-support", "case_count": 6, "personas": [ "Category Buyer", "Finance Director", "IT Asset Manager", "Procurement Manager" ], "prompts": [ "I need to identify all available discounts for our upcoming purchase orders: office supplies, IT equipment, and software licenses.", "Which IT equipment deals expire this week, and what should we lock in immediately?", "Model the office-supply bulk order strategy and the total savings projection for the quarter." ] }, "_catalog_kind": "solution", "_readiness": [ "Demo proven", "Conversation tested", "Synthetic data", "Production pattern" ] }, { "schema": "rapp-agent/1.0", "name": "@aibast-agents-library/care-gap-closure", "version": "1.1.0", "display_name": "Care Gap Closure Agent", "description": "Prioritizes synthetic Medicare Advantage care gaps before the HEDIS deadline: panel status, top gaps by revenue and reachability, diabetes risk tiers and barriers, a barrier-specific outreach strategy, campaign projection, and monitoring alerts, plus source-evidence views (gap_analysis, cohort_review, outreach_draft, quality_dashboard). Human clinical and quality review is required; it never determines eligibility, launches campaigns, or contacts patients.", "author": "AIBAST", "tags": [ "care-gaps", "quality-measures", "outreach-draft", "healthcare", "human-review" ], "category": "healthcare", "quality_tier": "verified", "requires_env": [], "dependencies": [ "@rapp/basic-agent" ], "_file": "agents/@aibast-agents-library/healthcare_stacks/care_gap_closure_stack/care_gap_closure_agent.py", "_size_kb": 19.6, "_lines": 477, "_sha256": "110f062d29d782d0a2eb8032b9c9c468112975997fc4d3d7d2eb0d3097998508", "_install_prefix": "care_gap_closure__", "_install_filename": "care_gap_closure__110f062d29d7_agent.py", "_stack": "care_gap_closure", "_stack_vertical": "healthcare", "_synthetic_data": true, "_added_at": "2026-03-18T13:23:12-04:00", "_solution": { "advertised_name": "Care Gap Closure Agent", "sharepoint_list_name": "Care Gap Closure Agent", "slot": 29, "executive_summary": "Automate quality gap analysis and targeted outreach to improve HEDIS performance, campaign ROI, and care gap closure efficiency.", "industries": [ "Healthcare" ], "personas": [ "Quality Managers", "Care Coordinators", "Clinical Operation Leads" ], "featured_tools": [ "Dynamics 365", "Microsoft Teams" ], "agent_requirements": [ "Microsoft Copilot Studio" ], "capabilities": [ "Analyze quality measures and surface high\u2011priority gaps", "Stratify outreach based on risk, condition severity, and engagement", "Identify common barriers that impacted care completion", "Launch targeted multi\u2011channel outreach campaigns with real\u2011time progress tracking" ], "outcomes": [ "Strengthened their HEDIS performance and Star Rating", "Improved care\u2011gap closure rates via targeted outreach", "Reduced admin workloads by automating patient identification and personalizing outreach", "Gained clear visibility into campaign progress to adjust strategies mid\u2011cycle" ], "customer_scenario": [ "Large volumes of open gaps across chronic and preventive measures", "Manual chart review and outreach processes that slowed progress on overdue visits and screenings", "Limited insight on access barriers and which patients were likely to respond", "No centralized view of gap\u2011closure progress across care teams" ], "slide_title": "Care Gap Closure Agent", "slide_summary": "Automates quality gap analysis and targeted outreach to improve HEDIS performance, campaign ROI, and care gap closure efficiency.", "scenario_name": "A regional health insurance provider", "challenge_intro": "A health insurance provider was at risk of missing quality thresholds due to:", "actions_intro": "The provider used the Care Gap Closure Agent to:", "outcomes_intro": "With the agent, the provider:", "opportunity_statements": [ "Analyze care gaps to surface priority measures enabling improved focus on high\u2011impact gaps", "Identify patient barriers and recommend interventions to boost outreach effectiveness", "Automate multi\u2011channel outreach campaigns and real-time monitoring to drive higher gap-closure rates" ], "module_name": null, "module_description": null, "source_verified": true, "sales_headline": "Turn aggregate quality evidence into a safer human review queue", "card_pitch": "Help quality and care-coordination teams inspect synthetic measure evidence, organize operational review cohorts, and draft unsent outreach while preserving clinical eligibility and privacy review.", "why_try": "The demo shows where an AI assistant can reduce review friction without pretending to decide who is eligible, overdue, high risk, or ready for outreach.", "customer_challenge_copy": "A regional health organization has large volumes of quality evidence, inconsistent barrier review, manual outreach preparation, and limited shared visibility\u2014but cannot compromise clinical eligibility rules or privacy controls.", "microsoft_ai_story": "Microsoft Copilot Studio provides the conversational experience. Dynamics 365 supports governed care-coordination work, and Microsoft Teams routes approved review tasks; the package itself uses aggregate synthetic evidence and sends nothing.", "business_value": [ "Makes aggregate source limitations and review queues easier for quality teams to see.", "Supports consistent operational cohorting without substituting an algorithm for clinical risk or eligibility review.", "Speeds outreach drafting while keeping consent, privacy, accessibility, and clinical approval explicit." ], "search_terms": [ "care gap closure", "quality measures", "HEDIS", "quality management", "care coordination", "outreach draft", "Dynamics 365 healthcare", "Microsoft Teams", "Copilot Studio", "population health" ], "journey_stage": "Govern", "blueprint_role": "Establishes a governed quality-evidence layer between aggregate source data and human clinical, eligibility, privacy, and outreach decisions.", "sample_prompts": [ { "label": "Quality Manager: Gap Analysis", "prompt": "Which synthetic quality measure has the largest evidence-review queue?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=care-gap-closure-gap-analysis" }, { "label": "Care Coordinator: Cohort Review", "prompt": "Organize the synthetic evidence-review cohorts without clinical risk scoring.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=care-gap-closure-cohort-review" }, { "label": "Care Coordinator: Outreach Draft", "prompt": "Draft outreach for the synthetic screening measure, but do not send it or state that anyone is eligible.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=care-gap-closure-outreach-draft" }, { "label": "Clinical Operations Lead: Quality Dashboard", "prompt": "Show the qualitative source-completeness dashboard and its limitations.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=care-gap-closure-quality-dashboard" }, { "label": "Quality Manager: Hedis Status", "prompt": "I need to prioritize care gaps for our Medicare Advantage population before the HEDIS reporting deadline. What's our current status?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=care-gap-closure-hedis-status" }, { "label": "Quality Manager: Top Gaps", "prompt": "Show me the top care gaps and which ones are most actionable.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=care-gap-closure-top-gaps" }, { "label": "Care Coordinator: Risk Stratification", "prompt": "Show me the diabetes cohort analysis with risk tiers and engagement barriers.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=care-gap-closure-risk-stratification" }, { "label": "Care Coordinator: Outreach Strategy", "prompt": "Design a barrier-specific outreach strategy for the diabetes gap patients.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=care-gap-closure-outreach-strategy" }, { "label": "Clinical Operations Lead: Campaign Projection", "prompt": "Show me the campaign deployment plan and projected impact.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=care-gap-closure-campaign-projection" }, { "label": "Clinical Operations Lead: Monitoring Plan", "prompt": "Configure monitoring for the campaign and alert me if performance drops.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=care-gap-closure-monitoring-plan" } ], "architecture": { "business_flow": [ "Quality managers and care coordinators", "Microsoft 365 Copilot or Copilot Studio", "Care Gap Closure Agent", "Dynamics 365 care-coordination workflow", "Microsoft Teams review and approval", "Quality and clinical reviewers" ], "capabilities": [ { "operation": "gap_analysis", "name": "Source-evidence gap analysis", "purpose": "Summarizes aggregate synthetic source counts and limitations without deciding measure eligibility." }, { "operation": "cohort_review", "name": "Aggregate cohort review", "purpose": "Organizes operational evidence-review cohorts without clinical risk scoring." }, { "operation": "outreach_draft", "name": "Unsent outreach draft", "purpose": "Drafts privacy-aware language but does not contact anyone." }, { "operation": "quality_dashboard", "name": "Qualitative quality dashboard", "purpose": "Shows source-completeness signals for quality and clinical validation." }, { "operation": "hedis_status", "name": "Hedis Status", "purpose": "Hedis Status" }, { "operation": "top_gaps", "name": "Top Gaps", "purpose": "Top Gaps" }, { "operation": "risk_stratification", "name": "Risk Stratification", "purpose": "Risk Stratification" }, { "operation": "outreach_strategy", "name": "Outreach Strategy", "purpose": "Outreach Strategy" }, { "operation": "campaign_projection", "name": "Campaign Projection", "purpose": "Campaign Projection" }, { "operation": "monitoring_plan", "name": "Monitoring Plan", "purpose": "Monitoring Plan" } ], "easy_mode": [ "Verify Power Platform CLI 2.9.3 or newer and authenticate to the target environment.", "Initialize a modern cli-copilot project with an approved publisher prefix.", "Give the Copilot Studio Architect the packaged GLOBAL-INSTRUCTIONS.md, two synthetic knowledge files, four operation skills, and this blueprint.", "Bind only dedicated read-only connection references; do not configure scheduling, outreach, submission, clinical-order, or record-write actions.", "Run pac copilot push and replay the four locked synthetic persona cases; stop before publish for human approval." ], "local_install_prompt": "You are GitHub Copilot running in Agent mode in VS Code. Install and open the AI BAST Care Gap Closure Agent from this deployment recipe: https://raw.githubusercontent.com/microsoft/aibast-agents-library/main/solutions/care-gap-closure/deployment.json. Own the setup without asking me to run terminal commands. Verify the source path and expected tool, install the source agent into the local Brainstem agents directory, start the Brainstem, verify /health, and run the recipe smoke prompt through /chat. Use only synthetic package data. Do not connect live patient systems, capture a global Brainstem, schedule, contact, submit, publish, or change any record. Ask me only for a supported browser authentication confirmation if required; otherwise stop on any failed verification and show the evidence. Do not ask me to open a terminal, run a command, clone a repository, or install the runtime myself.", "copilot_studio_prompt": "You are GitHub Copilot running in Agent mode in VS Code with the Microsoft Copilot Studio plugin installed. Prepare the AI BAST Care Gap Closure Agent from https://raw.githubusercontent.com/microsoft/aibast-agents-library/main/solutions/care-gap-closure/deployment.json. Ask only for the target Power Platform environment ID and publisher prefix. Initialize a cli-copilot project, author global instructions plus one skill per operation (gap_analysis, cohort_review, outreach_draft, quality_dashboard), upload only the two synthetic knowledge files, and create dedicated read-only connection references for Dynamics 365 care-coordination workflow, Microsoft Teams review and approval. Preserve the package's prohibition on diagnosis, treatment advice, eligibility or authorization decisions beyond source evidence, scheduling, outreach, submission, and record changes. Pull before push, run pac copilot push, replay all four locked cases, and stop before publish with validation and unresolved connection work. Stop before publish.", "required_connections": [ "Dynamics 365 care-coordination workflow", "Microsoft Teams review and approval" ], "manual_commands": [ "pac auth create", "pac copilot init --name \"Care Gap Closure Agent\" --publisher-prefix --authoring-mode cli-copilot --project-dir \"\" --environment \"\"", "pac connection list --environment \"\"", "pac copilot pull --project-dir \"\"", "pac copilot push --project-dir \"\"", "pac copilot publish --bot \"\" --environment \"\"" ], "acceptance_checks": [ "settings.mcs.yml preserves initialized identity, target environment, language, and cli-copilot authoring mode.", "Every advertised capability maps to the corresponding packaged skill and strict operation schema.", "Exactly two synthetic knowledge files are present and no live patient data is included.", "All four locked persona cases route to the expected agent and preserve evidence uncertainty and human review.", "No diagnosis, treatment, eligibility, authorization outcome, scheduling, outreach, submission, or record-write action is configured; publish remains separately approved." ], "hard_mode": [ "Run pac copilot init and verify settings.mcs.yml.", "Author the packaged global healthcare safety instructions and natural-language routing.", "Create one skill for each of the four implemented operations and upload exactly the two synthetic knowledge files.", "Create and verify dedicated read-only connection references with no mutating actions.", "Run pac copilot pull before push, inspect all generated YAML, replay the four locked cases, and stop before publication." ] }, "curated_copy": true, "source": "catalog", "origin": null, "onepager": "32. Care Gap Closure Agent one-pager.pptx", "demo_video": "32-Care Gap Closure Agent.mp4", "has_onepager": true, "has_demo_video": true, "promise_coverage": 75, "is_primary": true, "aliases": [], "package": { "slug": "care-gap-closure", "quest_url": "solutions/care-gap-closure/quest.html", "manual_tutorial_url": "solutions/care-gap-closure/manual-tutorial.html", "export_manifest_url": "solutions/care-gap-closure/export-manifest.json" } }, "_demo": { "slug": "care-gap-closure", "case_count": 10, "personas": [ "Care Coordinator", "Clinical Operations Lead", "Quality Manager" ], "prompts": [ "Which synthetic quality measure has the largest evidence-review queue?", "Organize the synthetic evidence-review cohorts without clinical risk scoring.", "Draft outreach for the synthetic screening measure, but do not send it or state that anyone is eligible." ] }, "_catalog_kind": "solution", "_readiness": [ "Demo proven", "Conversation tested", "Synthetic data", "Production pattern" ] }, { "schema": "rapp-agent/1.0", "name": "@aibast-agents-library/clinical-notes-summarizer", "version": "1.2.0", "display_name": "Clinical Notes Summarizer Agent", "description": "Handles synthetic clinical-note review: summarize encounters, list source-recorded medications, extract source-coded problem lists without confirming diagnoses, and summarize referral context; for pre-op clearance it assembles a patient snapshot, cardiopulmonary findings, a perioperative medication review, protocol-matched anesthesia considerations, ASA risk factors, re-evaluation criteria and a draft clearance note for physician signature. It requires clinician review and never diagnoses, recommends treatment, clears a patient, places a referral, schedules care, sends a message, or changes a record.", "author": "AIBAST", "tags": [ "clinical-notes", "source-summary", "medication-inventory", "healthcare", "clinician-review" ], "category": "healthcare", "quality_tier": "verified", "requires_env": [], "dependencies": [ "@rapp/basic-agent" ], "_file": "agents/@aibast-agents-library/healthcare_stacks/clinical_notes_summarizer_stack/clinical_notes_summarizer_agent.py", "_size_kb": 20.8, "_lines": 470, "_sha256": "2240a9e113288df4384b03876a2cfdb14189df251711b5c6e5d35cc4a3a70ac4", "_install_prefix": "clinical_notes_summarizer__", "_install_filename": "clinical_notes_summarizer__2240a9e11328_agent.py", "_stack": "clinical_notes_summarizer", "_stack_vertical": "healthcare", "_synthetic_data": true, "_added_at": "2026-03-18T13:23:12-04:00", "_solution": { "advertised_name": "Clinical Notes Summarizer Agent", "sharepoint_list_name": "Clinical Summary Agent", "slot": 47, "executive_summary": "Transform complex clinical histories into clear, actionable summaries for faster decision-making, better coordination, and safer care.", "industries": [ "Healthcare" ], "personas": [ "Primary care physicians", "Surgeons", "Anesthesia teams" ], "featured_tools": [ "Dynamics 365" ], "agent_requirements": [ "Microsoft Copilot Studio" ], "capabilities": [ "Automatically generate concise, clinically relevant patient summaries", "Surface key risks, comorbidities, and medication considerations", "Standardize documentation to support smoother surgical and anesthesia workflows", "Send summaries through EHR and secure messaging systems" ], "outcomes": [ "Completed chart reviews faster with consistent output", "Improved clinical safety by proactively flagging conditions and medications", "Reduced coordination delays across care teams with complete, ready\u2011to\u2011use summaries", "Improved physician productivity and rate of clearances delivered" ], "customer_scenario": [ "Clinicians spent too much time reviewing encounters, medications, and diagnostics", "Documentation quality varied, causing downstream delays", "High\u2011risk patients required additional coordination", "Critical risk factors and medication considerations were overlooked" ], "slide_title": "Clinical Notes Summarizer Agent", "slide_summary": "Transform complex clinical histories into clear, actionable summaries for faster decision\u2011making, better coordination, and safer care.", "scenario_name": "A multispecialty clinical practice", "challenge_intro": "A multispecialty practice struggled with slow, inconsistent chart reviews.", "actions_intro": "The practice used the Clinical Notes Summarizer Agent to:", "outcomes_intro": "With the Agent, the practice:", "opportunity_statements": [ "Synthesizes key information from patient charts to streamline clinical review and reduce manual effort", "Highlights relevant conditions, risks, and medication considerations to support confident clinical decisions", "Produces standardized summaries ready for surgical, anesthesia, and specialty workflows to improve team coordination" ], "module_name": null, "module_description": null, "source_verified": true, "sales_headline": "Turn long source notes into drafts clinicians can verify quickly", "card_pitch": "Give clinical teams consistent synthetic encounter, medication, problem-list, and referral-context drafts while keeping diagnosis, treatment, clearance, urgency, scheduling, and record changes with qualified humans.", "why_try": "The demo shows a transparent summarization pattern: every output separates source facts from clinical interpretation and reminds the reviewer to compare the draft with the authorized record.", "customer_challenge_copy": "A multispecialty practice faces slow and inconsistent chart review across encounters, medications, problem lists, and referrals, increasing coordination friction and the chance that source details are overlooked.", "microsoft_ai_story": "Microsoft Copilot Studio provides the conversational experience. Dynamics 365 or an approved healthcare data interface supplies governed clinical context; the package uses fictional records and has no clinical write path.", "business_value": [ "Makes source-fact review more consistent across encounter, medication, problem-list, and referral contexts.", "Reduces coordination friction by producing concise drafts with explicit source uncertainty.", "Protects clinical accountability by reserving diagnosis, treatment, clearance, urgency, and record changes for qualified reviewers." ], "search_terms": [ "clinical notes summarizer", "chart review", "encounter summary", "medication reconciliation", "problem list", "referral summary", "Dynamics 365 healthcare", "FHIR clinical records", "Copilot Studio", "clinician review" ], "journey_stage": "Activate", "blueprint_role": "Creates a governed source-summary layer that accelerates review without becoming a diagnostic, treatment, clearance, referral, scheduling, or record-write system.", "sample_prompts": [ { "label": "Summarize the source encounter", "prompt": "Summarize synthetic encounter SYN-ENC-001 using source facts only.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=clinical-notes-encounter" }, { "label": "Inventory recorded medications", "prompt": "List the source-recorded medications for SYN-ENC-001 for reconciliation.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=clinical-notes-medications" }, { "label": "Extract the problem list", "prompt": "Extract the source-coded problems for SYN-ENC-001 without confirming a diagnosis.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=clinical-notes-problems" }, { "label": "Review referral context", "prompt": "What referral context is recorded for SYN-ENC-001, and what action has not occurred?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=clinical-notes-referral" } ], "architecture": { "business_flow": [ "Physicians, surgeons, anesthesia teams, and clinical coordinators", "Microsoft 365 Copilot or Copilot Studio", "Clinical Notes Summarizer Agent", "Dynamics 365 or approved read-only healthcare record interface", "Qualified clinicians and pharmacists" ], "capabilities": [ { "operation": "encounter_summary", "name": "Source-grounded encounter summary", "purpose": "Extracts synthetic encounter facts without clinical inference." }, { "operation": "medication_inventory", "name": "Medication source inventory", "purpose": "Lists source-recorded synthetic medications for clinician or pharmacist reconciliation." }, { "operation": "problem_list_extract", "name": "Problem-list source extract", "purpose": "Extracts source-coded problems without confirming or changing a diagnosis." }, { "operation": "referral_context", "name": "Referral context extract", "purpose": "Summarizes source-recorded referral context without placing or scheduling it." } ], "easy_mode": [ "Verify Power Platform CLI 2.9.3 or newer and authenticate to the target environment.", "Initialize a modern cli-copilot project with an approved publisher prefix.", "Give the Copilot Studio Architect the packaged GLOBAL-INSTRUCTIONS.md, two synthetic knowledge files, four operation skills, and this blueprint.", "Bind only dedicated read-only connection references; do not configure scheduling, outreach, submission, clinical-order, or record-write actions.", "Run pac copilot push and replay the four locked synthetic persona cases; stop before publish for human approval." ], "local_install_prompt": "You are GitHub Copilot running in Agent mode in VS Code. Install and open the AI BAST Clinical Notes Summarizer Agent from this deployment recipe: https://raw.githubusercontent.com/microsoft/aibast-agents-library/main/solutions/clinical-notes-summarizer/deployment.json. Own the setup without asking me to run terminal commands. Verify the source path and expected tool, install the source agent into the local Brainstem agents directory, start the Brainstem, verify /health, and run the recipe smoke prompt through /chat. Use only synthetic package data. Do not connect live patient systems, capture a global Brainstem, schedule, contact, submit, publish, or change any record. Ask me only for a supported browser authentication confirmation if required; otherwise stop on any failed verification and show the evidence. Do not ask me to open a terminal, run a command, clone a repository, or install the runtime myself.", "copilot_studio_prompt": "You are GitHub Copilot running in Agent mode in VS Code with the Microsoft Copilot Studio plugin installed. Prepare the AI BAST Clinical Notes Summarizer Agent from https://raw.githubusercontent.com/microsoft/aibast-agents-library/main/solutions/clinical-notes-summarizer/deployment.json. Ask only for the target Power Platform environment ID and publisher prefix. Initialize a cli-copilot project, author global instructions plus one skill per operation (encounter_summary, medication_inventory, problem_list_extract, referral_context), upload only the two synthetic knowledge files, and create dedicated read-only connection references for Dynamics 365 or approved read-only healthcare record interface. Preserve the package's prohibition on diagnosis, treatment advice, eligibility or authorization decisions beyond source evidence, scheduling, outreach, submission, and record changes. Pull before push, run pac copilot push, replay all four locked cases, and stop before publish with validation and unresolved connection work. Stop before publish.", "required_connections": [ "Dynamics 365 or approved read-only healthcare record interface" ], "manual_commands": [ "pac auth create", "pac copilot init --name \"Clinical Notes Summarizer Agent\" --publisher-prefix --authoring-mode cli-copilot --project-dir \"\" --environment \"\"", "pac connection list --environment \"\"", "pac copilot pull --project-dir \"\"", "pac copilot push --project-dir \"\"", "pac copilot publish --bot \"\" --environment \"\"" ], "acceptance_checks": [ "settings.mcs.yml preserves initialized identity, target environment, language, and cli-copilot authoring mode.", "Every advertised capability maps to the corresponding packaged skill and strict operation schema.", "Exactly two synthetic knowledge files are present and no live patient data is included.", "All four locked persona cases route to the expected agent and preserve evidence uncertainty and human review.", "No diagnosis, treatment, eligibility, authorization outcome, scheduling, outreach, submission, or record-write action is configured; publish remains separately approved." ], "hard_mode": [ "Run pac copilot init and verify settings.mcs.yml.", "Author the packaged global healthcare safety instructions and natural-language routing.", "Create one skill for each of the four implemented operations and upload exactly the two synthetic knowledge files.", "Create and verify dedicated read-only connection references with no mutating actions.", "Run pac copilot pull before push, inspect all generated YAML, replay the four locked cases, and stop before publication." ] }, "curated_copy": true, "source": "catalog", "origin": null, "onepager": "49. Clinical Notes Summarizer Agent one-pager.pptx", "demo_video": "49-Clinical Notes Summarizer Agent.mp4", "has_onepager": true, "has_demo_video": true, "promise_coverage": 45, "is_primary": true, "aliases": [], "package": { "slug": "clinical-notes-summarizer", "quest_url": "solutions/clinical-notes-summarizer/quest.html", "manual_tutorial_url": "solutions/clinical-notes-summarizer/manual-tutorial.html", "export_manifest_url": "solutions/clinical-notes-summarizer/export-manifest.json" } }, "_demo": { "slug": "clinical-notes-summarizer", "case_count": 11, "personas": [ "Anesthesia Clinician", "Primary Care Physician", "Surgeon" ], "prompts": [ "Summarize synthetic encounter SYN-ENC-001 using source facts only.", "List the source-recorded medications for SYN-ENC-001 for reconciliation.", "Extract the source-coded problems for SYN-ENC-001 without confirming a diagnosis." ] }, "_catalog_kind": "solution", "_readiness": [ "Demo proven", "Conversation tested", "Synthetic data", "Production pattern" ] }, { "schema": "rapp-agent/1.0", "name": "@aibast-agents-library/patient-intake", "version": "1.2.0", "display_name": "Patient Intake and Scheduling Agent", "description": "Prepares synthetic new-patient intake, coverage evidence, provider availability, appointment request drafts, specialty intake packets, and reminder plans for patient-access staff; it never determines eligibility, books appointments, sends messages, or changes records.", "author": "AIBAST", "tags": [ "intake", "scheduling", "coverage-evidence", "reminders", "healthcare", "human-review" ], "category": "healthcare", "quality_tier": "verified", "requires_env": [], "dependencies": [ "@rapp/basic-agent" ], "_file": "agents/@aibast-agents-library/healthcare_stacks/patient_intake_stack/patient_intake_agent.py", "_size_kb": 21.0, "_lines": 465, "_sha256": "c94a5ac17c924c4f9e37a23dcd00a55eecb269394daee23810e82c9a53157496", "_install_prefix": "patient_intake__", "_install_filename": "patient_intake__c94a5ac17c92_agent.py", "_stack": "patient_intake", "_stack_vertical": "healthcare", "_synthetic_data": true, "_added_at": "2026-03-18T13:23:12-04:00", "_solution": { "advertised_name": "Patient Intake and Scheduling Agent", "sharepoint_list_name": "Patient Intake Agent", "slot": 28, "executive_summary": "Automate patient intake workflows to streamline operations, protect revenue from avoidable losses, and deliver a smoother patient experience.", "industries": [ "Healthcare" ], "personas": [ "Front Desk Staff", "Scheduling Coordinators", "Patient Access Reps" ], "featured_tools": [ "Dynamics 36", "SharePoint" ], "agent_requirements": [ "Microsoft Copilot Studio" ], "capabilities": [ "Verify insurance, referrals, and network status during patient intake", "Deliver tailored digital packets with all required pre\u2011visit forms", "Automate reminders across multiple scheduled touchpoints", "Ensure clinicians\u2019 availability and readiness through up\u2011to\u2011date scheduling data" ], "outcomes": [ "Reduced no\u2011shows with automated reminders, avoiding lost revenue", "Improved visit readiness by resolving eligibility and referral requirements upfront", "Streamlined digital intake, reducing manual data entry and wait\u2011times", "Freed staff capacity by automating routine scheduling, verification, and follow\u2011up tasks" ], "customer_scenario": [ "Manual scheduling led to long call queues and inconsistent appointment availability", "Eligibility checks occurred too late, leading to visit disruptions", "Paper intake packets slowed check\u2011ins and increased errors", "No\u2011show rates rose due to inconsistent reminder workflows" ], "slide_title": "Patient Intake and Scheduling Agent", "slide_summary": "Automates patient intake workflows to streamline operations, protect revenue from avoidable losses, and deliver a smoother patient experience.", "scenario_name": "A regional outpatient clinic", "challenge_intro": "A regional outpatient clinic faced rising patient\u2011access challenges:", "actions_intro": "The clinic used the Patient Intake and Scheduling Agent to:", "outcomes_intro": "With the agent, the clinic:", "opportunity_statements": [ "Automate intake checks and coordination to ensure patients are fully prepared for upcoming appointments", "Generate personalized digital intake packets to simplify documentation and reduce check\u2011in delays", "Run multi\u2011channel reminder and no\u2011show prevention workflows with rescheduling and waitlist automation" ], "module_name": null, "module_description": null, "source_verified": true, "sales_headline": "Make every visit-readiness handoff clear before staff take action", "card_pitch": "Give patient-access teams a safe, read-only view of synthetic intake completeness, recorded coverage evidence, candidate availability, and pre-visit follow-up\u2014without booking care or changing a record.", "why_try": "This demo turns a fragmented access workflow into four concise, reviewable handoffs while making the human and privacy boundaries visible in every answer.", "customer_challenge_copy": "A regional outpatient clinic is managing incomplete intake packets, late coverage follow-up, disconnected availability data, and inconsistent pre-visit handoffs without a single safe review surface.", "microsoft_ai_story": "Microsoft Copilot Studio provides the conversational front door. Dynamics 365 contributes approved patient-access workflow context, and SharePoint supplies governed intake documents; human staff retain eligibility, scheduling, communication, and record-change authority.", "business_value": [ "Improves consistency when staff review intake completeness and unresolved evidence.", "Clarifies coverage and availability follow-up without implying eligibility or booking.", "Reduces avoidable handoff ambiguity while preserving human approval and minimum-necessary privacy controls." ], "search_terms": [ "patient intake", "patient access", "intake readiness", "coverage evidence", "appointment availability", "pre-visit summary", "Dynamics 365 healthcare", "SharePoint forms", "Copilot Studio", "healthcare operations" ], "journey_stage": "Activate", "blueprint_role": "Creates a governed patient-access review layer that separates evidence preparation from eligibility, scheduling, communication, and record mutation.", "sample_prompts": [ { "label": "Front Desk Staff: Intake Readiness", "prompt": "What is still missing from the synthetic intake packet for Patient Alpha?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=patient-intake-intake-readiness" }, { "label": "Patient Access Representative: Coverage Evidence", "prompt": "What coverage evidence is recorded for Synthetic Patient Alpha, and what must staff still confirm?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=patient-intake-coverage-evidence" }, { "label": "Scheduling Coordinator: Appointment Availability", "prompt": "Show the candidate source slots for Clinician A without booking anything.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=patient-intake-appointment-availability" }, { "label": "Front Desk Staff: Pre Visit Summary", "prompt": "Prepare a pre-visit readiness summary for Synthetic Patient Alpha.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=patient-intake-pre-visit-summary" }, { "label": "Front Desk Staff: New Patient Intake", "prompt": "New patient Sarah Martinez called requesting a neurology consultation for chronic migraines within the next two weeks; start her registration.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=patient-intake-new-patient-intake" }, { "label": "Scheduling Coordinator: Appointment Hold Draft", "prompt": "Book Sarah Martinez for Tuesday at 2:30 and tell me which intake forms she needs.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=patient-intake-appointment-hold-draft" }, { "label": "Patient Access Representative: Intake Packet", "prompt": "Which forms go in Sarah Martinez's intake packet, and does it include HIPAA consent and medication history?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=patient-intake-intake-packet" }, { "label": "Scheduling Coordinator: Reminder Plan", "prompt": "Recommend automated reminders for Sarah Martinez's visit to reduce no-shows.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=patient-intake-reminder-plan" } ], "architecture": { "business_flow": [ "Patient-access staff and clinical operations reviewers", "Microsoft 365 Copilot or Copilot Studio", "Patient Intake and Scheduling Agent", "Dynamics 365 patient-access workflow", "SharePoint governed intake documents", "Authorized patient-access reviewers" ], "capabilities": [ { "operation": "intake_readiness", "name": "Intake readiness review", "purpose": "Summarizes present and missing synthetic intake evidence for staff confirmation." }, { "operation": "coverage_evidence", "name": "Coverage evidence review", "purpose": "Transcribes source-recorded synthetic coverage evidence without determining eligibility." }, { "operation": "appointment_availability", "name": "Appointment availability review", "purpose": "Shows candidate synthetic source slots without holding, booking, or changing an appointment." }, { "operation": "pre_visit_summary", "name": "Pre-visit readiness summary", "purpose": "Drafts a minimum-necessary readiness handoff for authorized review." }, { "operation": "new_patient_intake", "name": "New Patient Intake", "purpose": "New Patient Intake" }, { "operation": "appointment_hold_draft", "name": "Appointment Hold Draft", "purpose": "Appointment Hold Draft" }, { "operation": "intake_packet", "name": "Intake Packet", "purpose": "Intake Packet" }, { "operation": "reminder_plan", "name": "Reminder Plan", "purpose": "Reminder Plan" } ], "easy_mode": [ "Verify Power Platform CLI 2.9.3 or newer and authenticate to the target environment.", "Initialize a modern cli-copilot project with an approved publisher prefix.", "Give the Copilot Studio Architect the packaged GLOBAL-INSTRUCTIONS.md, two synthetic knowledge files, four operation skills, and this blueprint.", "Bind only dedicated read-only connection references; do not configure scheduling, outreach, submission, clinical-order, or record-write actions.", "Run pac copilot push and replay the four locked synthetic persona cases; stop before publish for human approval." ], "local_install_prompt": "You are GitHub Copilot running in Agent mode in VS Code. Install and open the AI BAST Patient Intake and Scheduling Agent from this deployment recipe: https://raw.githubusercontent.com/microsoft/aibast-agents-library/main/solutions/patient-intake/deployment.json. Own the setup without asking me to run terminal commands. Verify the source path and expected tool, install the source agent into the local Brainstem agents directory, start the Brainstem, verify /health, and run the recipe smoke prompt through /chat. Use only synthetic package data. Do not connect live patient systems, capture a global Brainstem, schedule, contact, submit, publish, or change any record. Ask me only for a supported browser authentication confirmation if required; otherwise stop on any failed verification and show the evidence. Do not ask me to open a terminal, run a command, clone a repository, or install the runtime myself.", "copilot_studio_prompt": "You are GitHub Copilot running in Agent mode in VS Code with the Microsoft Copilot Studio plugin installed. Prepare the AI BAST Patient Intake and Scheduling Agent from https://raw.githubusercontent.com/microsoft/aibast-agents-library/main/solutions/patient-intake/deployment.json. Ask only for the target Power Platform environment ID and publisher prefix. Initialize a cli-copilot project, author global instructions plus one skill per operation (intake_readiness, coverage_evidence, appointment_availability, pre_visit_summary), upload only the two synthetic knowledge files, and create dedicated read-only connection references for Dynamics 365 patient-access workflow, SharePoint governed intake documents. Preserve the package's prohibition on diagnosis, treatment advice, eligibility or authorization decisions beyond source evidence, scheduling, outreach, submission, and record changes. Pull before push, run pac copilot push, replay all four locked cases, and stop before publish with validation and unresolved connection work. Stop before publish.", "required_connections": [ "Dynamics 365 patient-access workflow", "SharePoint governed intake documents" ], "manual_commands": [ "pac auth create", "pac copilot init --name \"Patient Intake and Scheduling Agent\" --publisher-prefix --authoring-mode cli-copilot --project-dir \"\" --environment \"\"", "pac connection list --environment \"\"", "pac copilot pull --project-dir \"\"", "pac copilot push --project-dir \"\"", "pac copilot publish --bot \"\" --environment \"\"" ], "acceptance_checks": [ "settings.mcs.yml preserves initialized identity, target environment, language, and cli-copilot authoring mode.", "Every advertised capability maps to the corresponding packaged skill and strict operation schema.", "Exactly two synthetic knowledge files are present and no live patient data is included.", "All four locked persona cases route to the expected agent and preserve evidence uncertainty and human review.", "No diagnosis, treatment, eligibility, authorization outcome, scheduling, outreach, submission, or record-write action is configured; publish remains separately approved." ], "hard_mode": [ "Run pac copilot init and verify settings.mcs.yml.", "Author the packaged global healthcare safety instructions and natural-language routing.", "Create one skill for each of the four implemented operations and upload exactly the two synthetic knowledge files.", "Create and verify dedicated read-only connection references with no mutating actions.", "Run pac copilot pull before push, inspect all generated YAML, replay the four locked cases, and stop before publication." ] }, "curated_copy": true, "source": "catalog", "origin": null, "onepager": "31. Patient Intake and Scheduling Agent one-pager.pptx", "demo_video": "31-Patient Intake And Scheduling Agent.mp4", "has_onepager": true, "has_demo_video": true, "promise_coverage": 35, "is_primary": true, "aliases": [], "package": { "slug": "patient-intake", "quest_url": "solutions/patient-intake/quest.html", "manual_tutorial_url": "solutions/patient-intake/manual-tutorial.html", "export_manifest_url": "solutions/patient-intake/export-manifest.json" } }, "_demo": { "slug": "patient-intake", "case_count": 8, "personas": [ "Front Desk Staff", "Patient Access Representative", "Scheduling Coordinator" ], "prompts": [ "What is still missing from the synthetic intake packet for Patient Alpha?", "What coverage evidence is recorded for Synthetic Patient Alpha, and what must staff still confirm?", "Show the candidate source slots for Clinician A without booking anything." ] }, "_catalog_kind": "solution", "_readiness": [ "Demo proven", "Conversation tested", "Synthetic data", "Production pattern" ] }, { "schema": "rapp-agent/1.0", "name": "@aibast-agents-library/prior-authorization", "version": "1.2.0", "display_name": "Prior Authorization Agent", "description": "Assembles synthetic payer and clinical evidence for human utilization review: the patient request, payer-criteria match, a ready-to-submit packet, documentation strengths and appeal strategy, a proposed tracking plan, the authorization portfolio and denial reviews; it never predicts, grants, denies, submits, or changes an authorization and never sends a notification.", "author": "AIBAST", "tags": [ "prior-auth", "evidence-packet", "payer-criteria", "healthcare", "human-review" ], "category": "healthcare", "quality_tier": "verified", "requires_env": [], "dependencies": [ "@rapp/basic-agent" ], "_file": "agents/@aibast-agents-library/healthcare_stacks/prior_authorization_stack/prior_authorization_agent.py", "_size_kb": 22.2, "_lines": 497, "_sha256": "474c2429fae7cf94c1e625b6a99092e329b606e81dd29c58f1dbf1d5d4c3a05b", "_install_prefix": "prior_authorization__", "_install_filename": "prior_authorization__474c2429fae7_agent.py", "_stack": "prior_authorization", "_stack_vertical": "healthcare", "_synthetic_data": true, "_added_at": "2026-03-18T13:23:12-04:00", "_solution": { "advertised_name": "Prior Authorization Agent", "sharepoint_list_name": "Prior Authorization Agent", "slot": 45, "executive_summary": "Automate insurance approval workflows to accelerate authorization processes, improve documentation accuracy, and reduce care delays.", "industries": [ "Healthcare", "Financial Services" ], "personas": [ "Utilization Management Coordinators", "Nurse Case Managers", "Radiology Schedulers" ], "featured_tools": [ "Microsoft Teams" ], "agent_requirements": [ "Microsoft Copilot Studio" ], "capabilities": [ "Pull payer\u2011specific requirements and match them with EHR data", "Compile the full clinical evidence packet needed for medical review", "Submit requests directly to payer portals and send real\u2011time updates", "Flag cases requiring additional documentation and prepare appeal materials proactively" ], "outcomes": [ "Completed authorizations faster by relying on fewer manual steps", "Met payer criteria consistently, reducing denials", "Received timely notifications, enabling earlier appointment booking", "Gained clear visibility into authorization progress, reducing anxiety and delays" ], "customer_scenario": [ "Gathering documents and assessing payer criteria took too much time", "Incomplete submissions often led to denials and rework", "Providers and schedulers had limited visibility into authorization status", "Patients experienced uncertainty while waiting for approvals" ], "slide_title": "Prior Authorization Agent", "slide_summary": "Automate insurance approval workflows to accelerate authorization processes, improve documentation accuracy, and reduce care delays.", "scenario_name": "A multispecialty orthopedic practice", "challenge_intro": "An orthopedic clinic struggled with slow, inconsistent prior authorization workflows.", "actions_intro": "The clinic used the Prior Authorization Agent to:", "outcomes_intro": "With the agent in place, the clinic:", "opportunity_statements": [ "Performs intelligent payer requirement analysis and automated criteria matching to streamlines workflows", "Compiles comprehensive clinical evidence from the EHR to prevent denials and ensure first\u2011submission success", "Predicts approval likelihood and prepares automated appeal packages to protect revenue and reduce delays" ], "module_name": null, "module_description": null, "source_verified": true, "sales_headline": "Make authorization evidence reviewable without automating the decision", "card_pitch": "Give utilization teams a read-only synthetic inventory of request evidence, policy checklist context, recorded workflow state, and reconsideration material\u2014without predicting, submitting, granting, or denying authorization.", "why_try": "The demo makes a difficult governance boundary concrete: AI can organize evidence and uncertainty while a qualified reviewer retains policy interpretation, medical-necessity reasoning, and every outcome.", "customer_challenge_copy": "A multispecialty practice spends too much time gathering authorization evidence, checking policy requirements, tracking source states, and rebuilding reconsideration packets while clinicians and schedulers lack a safe shared view.", "microsoft_ai_story": "Microsoft Copilot Studio provides the conversational front door. Microsoft Teams coordinates utilization review, while approved healthcare and payer interfaces can supply read-only evidence; no live submission or outcome action is configured.", "business_value": [ "Improves consistency in criteria-to-evidence review and missing-document handoffs.", "Clarifies source-recorded workflow states without presenting them as agent decisions.", "Supports minimum-necessary reconsideration preparation while retaining qualified utilization review." ], "search_terms": [ "prior authorization", "utilization management", "payer criteria", "clinical evidence packet", "authorization status", "reconsideration evidence", "Microsoft Teams healthcare", "FHIR", "Copilot Studio", "medical necessity review" ], "journey_stage": "Govern", "blueprint_role": "Creates the evidence-preparation and governance layer between clinical source systems, payer policy, utilization review, and authorization outcomes.", "sample_prompts": [ { "label": "Inventory request evidence", "prompt": "What evidence is present or missing for synthetic request SYN-AUTH-001?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=prior-auth-evidence" }, { "label": "Crosswalk the policy checklist", "prompt": "Show the synthetic criteria checklist for SYN-AUTH-001 without deciding medical necessity.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=prior-auth-criteria" }, { "label": "Read the recorded status", "prompt": "What workflow state is recorded for SYN-AUTH-001, and what does it not mean?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=prior-auth-status" }, { "label": "Prepare review material", "prompt": "Prepare a minimum-necessary reconsideration evidence outline for SYN-AUTH-001.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=prior-auth-reconsideration" } ], "architecture": { "business_flow": [ "Utilization coordinators, nurse case managers, and schedulers", "Microsoft 365 Copilot or Copilot Studio", "Prior Authorization Agent", "Approved read-only EHR or FHIR evidence source", "Approved payer policy source", "Microsoft Teams utilization-review coordination", "Qualified utilization reviewers" ], "capabilities": [ { "operation": "request_evidence", "name": "Request evidence inventory", "purpose": "Inventories synthetic request evidence without submitting or deciding the request." }, { "operation": "criteria_evidence", "name": "Criteria-to-evidence crosswalk", "purpose": "Presents synthetic checklist items for qualified utilization review." }, { "operation": "status_summary", "name": "Source-recorded status summary", "purpose": "Transcribes a synthetic workflow state without making an authorization outcome." }, { "operation": "appeal_evidence_packet", "name": "Reconsideration evidence draft", "purpose": "Prepares a minimum-necessary outline without filing or recommending an appeal." } ], "easy_mode": [ "Verify Power Platform CLI 2.9.3 or newer and authenticate to the target environment.", "Initialize a modern cli-copilot project with an approved publisher prefix.", "Give the Copilot Studio Architect the packaged GLOBAL-INSTRUCTIONS.md, two synthetic knowledge files, four operation skills, and this blueprint.", "Bind only dedicated read-only connection references; do not configure scheduling, outreach, submission, clinical-order, or record-write actions.", "Run pac copilot push and replay the four locked synthetic persona cases; stop before publish for human approval." ], "local_install_prompt": "You are GitHub Copilot running in Agent mode in VS Code. Install and open the AI BAST Prior Authorization Agent from this deployment recipe: https://raw.githubusercontent.com/microsoft/aibast-agents-library/main/solutions/prior-authorization/deployment.json. Own the setup without asking me to run terminal commands. Verify the source path and expected tool, install the source agent into the local Brainstem agents directory, start the Brainstem, verify /health, and run the recipe smoke prompt through /chat. Use only synthetic package data. Do not connect live patient systems, capture a global Brainstem, schedule, contact, submit, publish, or change any record. Ask me only for a supported browser authentication confirmation if required; otherwise stop on any failed verification and show the evidence. Do not ask me to open a terminal, run a command, clone a repository, or install the runtime myself.", "copilot_studio_prompt": "You are GitHub Copilot running in Agent mode in VS Code with the Microsoft Copilot Studio plugin installed. Prepare the AI BAST Prior Authorization Agent from https://raw.githubusercontent.com/microsoft/aibast-agents-library/main/solutions/prior-authorization/deployment.json. Ask only for the target Power Platform environment ID and publisher prefix. Initialize a cli-copilot project, author global instructions plus one skill per operation (request_evidence, criteria_evidence, status_summary, appeal_evidence_packet), upload only the two synthetic knowledge files, and create dedicated read-only connection references for Approved read-only EHR or FHIR evidence source, Approved payer policy source, Microsoft Teams utilization-review coordination. Preserve the package's prohibition on diagnosis, treatment advice, eligibility or authorization decisions beyond source evidence, scheduling, outreach, submission, and record changes. Pull before push, run pac copilot push, replay all four locked cases, and stop before publish with validation and unresolved connection work. Stop before publish.", "required_connections": [ "Approved read-only EHR or FHIR evidence source", "Approved payer policy source", "Microsoft Teams utilization-review coordination" ], "manual_commands": [ "pac auth create", "pac copilot init --name \"Prior Authorization Agent\" --publisher-prefix --authoring-mode cli-copilot --project-dir \"\" --environment \"\"", "pac connection list --environment \"\"", "pac copilot pull --project-dir \"\"", "pac copilot push --project-dir \"\"", "pac copilot publish --bot \"\" --environment \"\"" ], "acceptance_checks": [ "settings.mcs.yml preserves initialized identity, target environment, language, and cli-copilot authoring mode.", "Every advertised capability maps to the corresponding packaged skill and strict operation schema.", "Exactly two synthetic knowledge files are present and no live patient data is included.", "All four locked persona cases route to the expected agent and preserve evidence uncertainty and human review.", "No diagnosis, treatment, eligibility, authorization outcome, scheduling, outreach, submission, or record-write action is configured; publish remains separately approved." ], "hard_mode": [ "Run pac copilot init and verify settings.mcs.yml.", "Author the packaged global healthcare safety instructions and natural-language routing.", "Create one skill for each of the four implemented operations and upload exactly the two synthetic knowledge files.", "Create and verify dedicated read-only connection references with no mutating actions.", "Run pac copilot pull before push, inspect all generated YAML, replay the four locked cases, and stop before publication." ] }, "curated_copy": true, "source": "catalog", "origin": null, "onepager": "47. Prior Authorization Agent one-pager.pptx", "demo_video": "47-Prior Authorization Automation Agent.mp4", "has_onepager": true, "has_demo_video": true, "promise_coverage": 40, "is_primary": true, "aliases": [], "package": { "slug": "prior-authorization", "quest_url": "solutions/prior-authorization/quest.html", "manual_tutorial_url": "solutions/prior-authorization/manual-tutorial.html", "export_manifest_url": "solutions/prior-authorization/export-manifest.json" } }, "_demo": { "slug": "prior-authorization", "case_count": 11, "personas": [ "Nurse Case Manager", "Radiology Scheduler", "Utilization Management Coordinator" ], "prompts": [ "What evidence is present or missing for synthetic request SYN-AUTH-001?", "Show the synthetic criteria checklist for SYN-AUTH-001 without deciding medical necessity.", "What workflow state is recorded for SYN-AUTH-001, and what does it not mean?" ] }, "_catalog_kind": "solution", "_readiness": [ "Demo proven", "Conversation tested", "Synthetic data", "Production pattern" ] }, { "schema": "rapp-agent/1.0", "name": "@aibast-agents-library/tiered-huddle-board", "version": "1.0.0", "display_name": "Tiered Huddle Board Agent", "description": "Run a hospital's daily tiered huddles from one live board built on a synthetic snapshot: every lane from Tier 1 through the Executive tier with each issue's clock against its tier's target, what is past target and what is waiting on the person asking, an escalation draft that restarts the clock at the next tier's target with the reason the next tier reads first and the people to page by name, the closure check that keeps mitigated items open until the person who raised them confirms the fix held, and the decisions the leadership huddle needs today. The agent never escalates, pages, closes or changes an issue; an authorized person confirms every step through future authenticated tools.", "author": "AIBAST", "tags": [ "healthcare", "tiered-huddles", "nursing-operations", "escalation", "hospital-operations" ], "category": "healthcare", "quality_tier": "verified", "requires_env": [], "dependencies": [ "@rapp/basic-agent" ], "_file": "agents/@aibast-agents-library/healthcare_stacks/tiered_huddle_board_stack/tiered_huddle_board_agent.py", "_size_kb": 29.8, "_lines": 564, "_sha256": "f1d3530dce1bbe426732848af126e76cc4eddbe565d8103d2ddb8197acd97288", "_install_prefix": "tiered_huddle_board__", "_install_filename": "tiered_huddle_board__f1d3530dce1b_agent.py", "_stack": "tiered_huddle_board", "_stack_vertical": "healthcare", "_synthetic_data": true, "_added_at": "2026-10-28T18:25:15-07:00", "_solution": { "advertised_name": "Tiered Huddle Board Agent", "sharepoint_list_name": "Tiered Huddle Board Agent", "slot": 71, "executive_summary": "Help a hospital run its daily tiered huddles from one live board: every lane and clock, what is past target and waiting on the person asking, escalations drafted with the restarted clock, the reason and the people to page, mitigated issues kept open until the person who raised them confirms the fix held, and the decisions the leadership huddle needs today.", "industries": [ "Healthcare", "Hospitals & Health Systems" ], "personas": [ "Nurse Manager", "Department Director", "Chief Nursing Officer" ], "featured_tools": [ "Microsoft Dataverse", "Microsoft Teams", "Power Apps" ], "agent_requirements": [ "Microsoft Copilot Studio" ], "capabilities": [ "Show every lane, clock and item waiting on the person asking", "Draft each escalation with its reason, new clock and pages", "Keep mitigated items open until the originator confirms", "Bring leadership every decision with its reason and clock" ], "outcomes": [ "Ran every huddle tier from one live board", "Escalated unresolved issues faster, with nothing lost", "Closed issues only when the fix had truly held", "Gave the 10:30 huddle a clear list of decisions" ], "customer_scenario": [ "Tiers, targets and paging lists lived in a paper binder", "Nurse managers could not see the board between patients", "Escalations lost their reason and clock on the way up", "Mitigated issues were closed before the fix had held" ], "slide_title": "Tiered Huddle Board Agent", "slide_summary": "Gives every huddle tier one live board, drafts escalations for a person to confirm, and keeps issues open until the fix holds.", "scenario_name": "A community hospital", "challenge_intro": "A community hospital struggled to run its daily tiered huddles:", "actions_intro": "The hospital leveraged the Tiered Huddle Board Agent to:", "outcomes_intro": "With the agent\u2019s help, the hospital:", "opportunity_statements": [ "Show the whole board with every lane, clock and item waiting on the person asking, from Copilot between patients", "Escalate with one confirmation, restarting the clock at the next tier with the reason and the right people paged", "Keep mitigated items open until the person who raised them confirms the fix held, and show leadership what to decide" ], "module_name": null, "module_description": null, "source_verified": true, "sales_headline": "Run every tier of the daily huddle from one live board", "card_pitch": "Give nurse managers, department directors and hospital leaders one conversational view of the huddle board: every lane and clock, what is past target, escalations drafted with the reason and the people to page, and mitigated issues held open until the fix is confirmed.", "why_try": "The demo lets a nurse manager ask between patients what is past target and waiting on her, then draft an escalation that a person confirms, and shows the 10:30 leadership huddle exactly what needs a decision.", "customer_challenge_copy": "Hospitals run tiered huddles from binders and whiteboards, so escalations lose their reason and clock on the way up, the right people are paged late, and issues are closed before anyone confirms the fix held.", "microsoft_ai_story": "Copilot Studio gives each tier a conversational view of the board in Microsoft Teams or Microsoft 365 Copilot; Microsoft Dataverse holds the issues, tiers, clocks and paging roster behind a Power Apps board, and Teams carries the pages once a person confirms the escalation.", "business_value": [ "Shows every lane and clock, so past-target items surface before the huddle.", "Escalates with the reason, a restarted clock and the right people paged after one confirmation.", "Keeps mitigated issues open until the person who raised them confirms the fix held." ], "search_terms": [ "Chief Nursing Officer", "Department Director", "Microsoft Copilot Studio", "Microsoft Dataverse", "Microsoft Teams", "Nurse Manager", "Power Apps", "Tiered Huddle Board Agent", "board_view", "closure_check", "daily safety huddle", "escalation", "escalation_draft", "hospital operations", "huddle board", "leadership huddle", "leadership_decisions", "nursing operations", "paging", "tiered huddle" ], "journey_stage": "Optimize", "blueprint_role": "Separates board visibility and escalation preparation from protected escalation, paging, closure and leadership decisions.", "sample_prompts": [ { "label": "Nurse Manager: Whole board view", "prompt": "I'm between patients. Show me the huddle board: what's past target and what's waiting on me?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=tiered-huddle-board-board-view" }, { "label": "Nurse Manager: Escalation draft", "prompt": "4 West can't fix the IV pump problem at our level. Get the escalation ready for the next tier, but don't page anyone until I confirm.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=tiered-huddle-board-escalation-draft" }, { "label": "Nurse Manager: Closure check", "prompt": "Biomed fixed the bed alarm in room 412 yesterday. Can we take it off the board?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=tiered-huddle-board-closure-check" }, { "label": "Chief Nursing Officer: Leadership huddle decisions", "prompt": "For the 10:30 leadership huddle, what needs a decision from us today?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=tiered-huddle-board-leadership-decisions" } ], "architecture": { "business_flow": [ "Nurse managers, department directors and hospital leaders", "Microsoft Teams or Microsoft 365 Copilot", "Tiered Huddle Board Agent", "Microsoft Dataverse huddle board: issues, tiers and clocks", "Paging roster by tier and issue category", "Authorized escalation and closure step on the Power Apps board" ], "capabilities": [ { "operation": "board_view", "name": "Whole board view", "purpose": "Use when a nurse manager, charge nurse or director asks for the huddle board, every lane and clock, what is past target, or what is waiting on them." }, { "operation": "escalation_draft", "name": "Escalation draft", "purpose": "Use when someone says the unit or department cannot fix an issue and asks to escalate it or who would be paged; returns an escalation draft that is never sent." }, { "operation": "closure_check", "name": "Closure check", "purpose": "Use when someone asks whether a fixed or mitigated issue can be closed; mitigated is not closed until the person who raised it confirms the fix held." }, { "operation": "leadership_decisions", "name": "Leadership huddle decisions", "purpose": "Use when a leader asks what the 10:30 leadership huddle needs to decide today, with the reason and clock for each item." } ], "easy_mode": [ "Verify Power Platform CLI 2.9.3 or newer and authenticate to the approved target environment.", "Initialize a modern cli-copilot project with an approved publisher prefix and preserve the generated identity settings.", "Upload the two packaged synthetic knowledge files and every operation-specific SKILL.md.", "Bind only approved least-privilege production connections after security, privacy, and business-owner review.", "Push the Draft agent, replay every locked persona case, and stop before publish for explicit approval." ], "local_install_prompt": "Install Tiered Huddle Board Agent from https://raw.githubusercontent.com/microsoft/aibast-agents-library/main/solutions/tiered-huddle-board/deployment.json, verify the expected tool and smoke case, and preserve the synthetic read-only boundary. Do not ask me to open a terminal. Do not connect hospital systems or execute external actions.", "copilot_studio_prompt": "Using the Microsoft Copilot Studio plugin, initialize a Draft agent for Tiered Huddle Board Agent from https://raw.githubusercontent.com/microsoft/aibast-agents-library/main/solutions/tiered-huddle-board/deployment.json. Upload the two synthetic knowledge files and 4 skills, replay every locked case, and Stop before publish. Do not bind production connections without explicit customer approval.", "required_connections": [ "Use only approved, least-privilege production connections; the packaged demonstration uses synthetic knowledge files.", "Microsoft Dataverse huddle board tables: issues, tiers, clocks and paging roster (read-only)", "Microsoft Teams for escalation pages, sent only by the board's authorized escalation step", "Microsoft Entra ID user identity for 'waiting on me' and the issue-owner confirmation" ], "manual_commands": [ "pac auth create", "pac copilot init --name \"Tiered Huddle Board Agent\" --publisher-prefix --authoring-mode cli-copilot --project-dir \"\" --environment \"\"", "pac connection list --environment \"\"", "pac copilot pull --project-dir \"\"", "pac copilot push --project-dir \"\"", "pac copilot publish --bot \"\" --environment \"\"" ], "acceptance_checks": [ "Every exposed operation has one persona-language case and one SKILL.md.", "Exactly two synthetic knowledge files are present.", "Responses state the relevant human confirmation or review boundary.", "No operation claims an escalation, page, closure, decision or record change.", "Publishing remains a separate user-approved step." ], "hard_mode": [ "Run pac copilot init and verify settings.mcs.yml before authoring components.", "Author global instructions that preserve the packaged synthetic-only and human-confirmation boundaries.", "Add the two knowledge files and one skill for every exposed operation.", "Create dedicated connection references only for approved systems and verify least-privilege access.", "Pull before push, inspect every generated file, replay every locked case, and stop before publish." ] }, "curated_copy": true, "source": "catalog", "origin": null, "onepager": "71. Tiered Huddle Board Agent one-pager.pptx", "demo_video": null, "has_onepager": true, "has_demo_video": false, "promise_coverage": 100, "is_primary": true, "aliases": [], "package": { "slug": "tiered-huddle-board", "quest_url": "solutions/tiered-huddle-board/quest.html", "manual_tutorial_url": "solutions/tiered-huddle-board/manual-tutorial.html", "export_manifest_url": "solutions/tiered-huddle-board/export-manifest.json" } }, "_demo": { "slug": "tiered-huddle-board", "case_count": 4, "personas": [ "Chief Nursing Officer", "Nurse Manager" ], "prompts": [ "I'm between patients. Show me the huddle board: what's past target and what's waiting on me?", "4 West can't fix the IV pump problem at our level. Get the escalation ready for the next tier, but don't page anyone until I confirm.", "Biomed fixed the bed alarm in room 412 yesterday. Can we take it off the board?" ] }, "_catalog_kind": "solution", "_readiness": [ "Demo proven", "Conversation tested", "Synthetic data", "Production pattern" ] }, { "schema": "rapp-agent/1.0", "name": "@aibast-agents-library/ask-hr", "version": "1.0.0", "display_name": "Ask HR Agent", "description": "Provide self-service HR inquiry handling that transforms the process from a manual ticket-based system to intelligent, automated resolutions.", "author": "AIBAST", "tags": [ "hr", "human-resources", "benefits", "time-off", "employee-self-service" ], "category": "human_resources", "quality_tier": "verified", "requires_env": [], "dependencies": [ "@rapp/basic-agent" ], "_file": "agents/@aibast-agents-library/human_resources_stacks/ask_hr_stack/ask_hr_agent.py", "_size_kb": 27.6, "_lines": 578, "_sha256": "11b66d0eb4bb9d1ed89663bea3722fa9a69fc4f2e26ca461bf99ad1634ef951e", "_install_prefix": "ask_hr__", "_install_filename": "ask_hr__11b66d0eb4bb_agent.py", "_stack": "ask_hr", "_stack_vertical": "human_resources", "_synthetic_data": true, "_added_at": "2026-03-18T13:23:12-04:00", "_solution": { "advertised_name": "Ask HR Agent", "sharepoint_list_name": "HR Self-Service Agent", "slot": 46, "executive_summary": "Provide self-service HR inquiry handling that transforms the process from a manual ticket-based system to intelligent, automated resolutions.", "industries": [ "Cross-Industry" ], "personas": [ "Employees", "Managers", "HR Operations Staff" ], "featured_tools": [ "Outlook", "Microsoft Teams" ], "agent_requirements": [ "Microsoft Copilot Studio" ], "capabilities": [ "Give employees instant, personalized answers based on their profile, location, and benefits", "Surface relevant policy guidance and upcoming deadlines", "Automate time\u2011off requests, form pre\u2011population, and approval routing", "Reduce dependency on HR staff for routine questions and tasks" ], "outcomes": [ "Resolved common HR inquiries instantly, without tickets", "Eliminated bottlenecks in time\u2011off management with automated routing and notifications", "Reduced HR workloads, freeing up time for higher-value work", "Improved the employee experience through clear, consistent guidance" ], "customer_scenario": [ "Employees waited for answers to various policy questions", "HR teams spent significant time on repetitive, low\u2011complexity inquiries", "Manual time\u2011off submissions and approval workflows led to delays", "Employees lacked clarity on enrollment windows, balances, and required actions" ], "slide_title": "Ask HR Agent", "slide_summary": "Provide self\u2011service HR inquiry handling that transforms the process from a manual ticket-based system to intelligent, automated resolutions.", "scenario_name": "A global technology company", "challenge_intro": "A fast\u2011growing technology company struggled with rising HR inquiry volume.", "actions_intro": "The company deployed the Ask HR Agent to:", "outcomes_intro": "With the agent in place, the organization:", "opportunity_statements": [ "Provides personalized, policy\u2011based answers to reduce inquiry volume via 24/7 self-service resolution", "Guides employees through HR tasks like benefit plan selection to support better employee experiences", "Automates core HR transactions like time\u2011off requests to reduce operational costs and increase HR team productivity" ], "module_name": null, "module_description": null, "source_verified": true, "sales_headline": "Answer routine HR questions without turning guidance into a decision", "card_pitch": "Give employees a privacy-conscious self-service front door for fictional balances and published policy examples, while keeping eligibility, sensitive circumstances, submissions, and HR decisions with authorized people and systems.", "why_try": "The demo lets an employee ask a natural question and receive a source-backed, reviewable answer with the approval boundary visible in the response.", "customer_challenge_copy": "Employees wait for routine policy answers and HR teams repeat low-complexity guidance, yet personalization can expose sensitive data or overstep into eligibility decisions.", "microsoft_ai_story": "Copilot Studio and Microsoft Teams provide the employee experience; approved HRIS, benefits, and policy sources can be bound with identity, row-level security, data minimization, and audit controls.", "business_value": [ "Improves access to consistent policy guidance.", "Reduces repetitive explanation work for HR operations.", "Preserves privacy and human authority for individualized decisions." ], "search_terms": [ "Ask HR Agent", "Employee", "HR Operations Specialist", "Manager", "Microsoft Copilot Studio", "Microsoft Teams", "benefits_summary", "health_insurance", "leave_balance", "parental_leave", "remote_work", "submit_time_off" ], "journey_stage": "Empower", "blueprint_role": "Establishes a safe self-service layer that separates policy explanation and draft preparation from protected HR transactions and decisions.", "sample_prompts": [ { "label": "Employee: Leave-balance explanation", "prompt": "What does my fictional leave snapshot show, and what notice rules should I review?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=ask-hr-leave-balance" }, { "label": "Employee: Time-off draft preview", "prompt": "Preview five vacation days for September 14 through September 18; do not submit anything.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=ask-hr-submit-time-off" }, { "label": "Employee: Parental-leave policy guidance", "prompt": "Explain the published parental-leave example without deciding whether I qualify or asking for family details.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=ask-hr-parental-leave" }, { "label": "Employee: Health-plan policy explanation", "prompt": "Summarize the fictional health-plan snapshot and tell me what the benefits administrator must verify.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=ask-hr-health-insurance" }, { "label": "Manager: Remote-work policy guidance", "prompt": "What does the sample remote-work policy say, without inferring why an employee asked?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=ask-hr-remote-work" }, { "label": "HR Operations Specialist: Benefits snapshot explanation", "prompt": "Show the fictional benefits snapshot without estimating compensation or making an eligibility decision.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=ask-hr-benefits-summary" } ], "architecture": { "business_flow": [ "Employees, managers, and HR operations", "Microsoft Teams or Microsoft 365 Copilot", "Ask HR Agent", "Approved HRIS and benefits systems", "Policy knowledge source", "Authorized HR reviewer and workflow" ], "capabilities": [ { "operation": "leave_balance", "name": "Leave-balance explanation", "purpose": "Use when an employee asks for the fictional balance snapshot and published time-off rules." }, { "operation": "submit_time_off", "name": "Time-off draft preview", "purpose": "Use when an employee wants to preview dates and projected balance before using the approved submission workflow." }, { "operation": "parental_leave", "name": "Parental-leave policy guidance", "purpose": "Use for published leave-policy questions without inferring a family or medical circumstance or deciding eligibility." }, { "operation": "health_insurance", "name": "Health-plan policy explanation", "purpose": "Use when an employee asks what the fictional plan snapshot says and what must be verified with benefits staff." }, { "operation": "remote_work", "name": "Remote-work policy guidance", "purpose": "Use when an employee or manager asks for published remote-work rules without inferring a sensitive circumstance." }, { "operation": "benefits_summary", "name": "Benefits snapshot explanation", "purpose": "Use when HR operations needs a concise fictional profile summary without estimating compensation or deciding eligibility." } ], "easy_mode": [ "Verify Power Platform CLI 2.9.3 or newer and authenticate to the approved target environment.", "Initialize a modern cli-copilot project with an approved publisher prefix and preserve the generated identity settings.", "Upload the two packaged synthetic knowledge files and every operation-specific SKILL.md.", "Bind only approved least-privilege production connections after security, privacy, and business-owner review.", "Push the Draft agent, replay every locked persona case, and stop before publish for explicit approval." ], "local_install_prompt": "Install Ask HR Agent from https://raw.githubusercontent.com/microsoft/aibast-agents-library/main/solutions/ask-hr/deployment.json, verify the expected tool and smoke case, and preserve the synthetic read-only boundary. Do not ask me to open a terminal. Do not connect customer systems or execute external actions.", "copilot_studio_prompt": "Using the Microsoft Copilot Studio plugin, initialize a Draft agent for Ask HR Agent from https://raw.githubusercontent.com/microsoft/aibast-agents-library/main/solutions/ask-hr/deployment.json. Upload the two synthetic knowledge files and 6 skills, replay every locked case, and Stop before publish. Do not bind production connections without explicit customer approval.", "required_connections": [ "Use only approved, least-privilege production connections; the packaged demonstration uses synthetic knowledge files." ], "manual_commands": [ "pac auth create", "pac copilot init --name \"Ask HR Agent\" --publisher-prefix --authoring-mode cli-copilot --project-dir \"\" --environment \"\"", "pac connection list --environment \"\"", "pac copilot pull --project-dir \"\"", "pac copilot push --project-dir \"\"", "pac copilot publish --bot \"\" --environment \"\"" ], "acceptance_checks": [ "Every exposed operation has one persona-language case and one SKILL.md.", "Exactly two synthetic knowledge files are present.", "Responses state the relevant human approval or review boundary.", "No operation claims an external write, communication, commitment, or protected decision.", "Publishing remains a separate user-approved step." ], "hard_mode": [ "Run pac copilot init and verify settings.mcs.yml before authoring components.", "Author global instructions that preserve the packaged synthetic-only and human-approval boundaries.", "Add the two knowledge files and one skill for every exposed operation.", "Create dedicated connection references only for approved systems and verify least-privilege access.", "Pull before push, inspect every generated file, replay every locked case, and stop before publish." ] }, "curated_copy": true, "source": "catalog", "origin": null, "onepager": "48. Ask HR Agent one-pager.pptx", "demo_video": "48-Ask HR Agent.mp4", "has_onepager": true, "has_demo_video": true, "promise_coverage": 85, "is_primary": true, "aliases": [], "package": { "slug": "ask-hr", "quest_url": "solutions/ask-hr/quest.html", "manual_tutorial_url": "solutions/ask-hr/manual-tutorial.html", "export_manifest_url": "solutions/ask-hr/export-manifest.json" } }, "_demo": { "slug": "ask-hr", "case_count": 8, "personas": [ "Employee", "HR Operations Specialist", "Manager" ], "prompts": [ "What does my fictional leave snapshot show, and what notice rules should I review?", "Preview five vacation days for September 14 through September 18; do not submit anything.", "Explain the published parental-leave example without deciding whether I qualify or asking for family details." ] }, "_catalog_kind": "solution", "_readiness": [ "Demo proven", "Conversation tested", "Synthetic data", "Production pattern" ] }, { "schema": "rapp-agent/1.0", "name": "@aibast-agents-library/benefits-enrollment-assistant", "version": "1.0.0", "display_name": "Benefits Enrollment Agent", "description": "Guide employees through open enrollment and life-event benefit changes by explaining the enrollment window and change rules, comparing plan costs for their coverage tier, checking whether their providers are in network, tracking required documents, and preparing a draft election, while eligibility decisions and submissions stay with the employee and benefits staff.", "author": "AIBAST", "tags": [ "hr", "benefits", "open-enrollment", "life-events", "human-resources" ], "category": "human_resources", "quality_tier": "verified", "requires_env": [], "dependencies": [ "@rapp/basic-agent" ], "_file": "agents/@aibast-agents-library/human_resources_stacks/benefits_enrollment_assistant_stack/benefits_enrollment_assistant_agent.py", "_size_kb": 22.2, "_lines": 432, "_sha256": "61bfddee0e6d9127da21b9fea2888fe15777a6364e3d9e5fc919eaf9a2ad653b", "_install_prefix": "benefits_enrollment_assistant__", "_install_filename": "benefits_enrollment_assistant__61bfddee0e6d_agent.py", "_stack": "benefits_enrollment_assistant", "_stack_vertical": "human_resources", "_synthetic_data": true, "_added_at": "2026-10-08T10:30:56-04:00", "_solution": { "advertised_name": "Benefits Enrollment Agent", "sharepoint_list_name": "Benefits Enrollment Agent", "slot": 63, "executive_summary": "Guide employees through open enrollment and life-event benefit changes by explaining the enrollment window and change rules, comparing plan costs for their coverage tier, checking whether their providers are in network, tracking required documents, and preparing a draft election, while eligibility decisions and submissions stay with the employee and benefits staff.", "industries": [ "Cross-Industry" ], "personas": [ "Employee", "Benefits Specialist", "HR Operations Staff" ], "featured_tools": [ "Microsoft Teams", "Microsoft 365 Copilot" ], "agent_requirements": [ "Microsoft Copilot Studio" ], "capabilities": [ "Explains the enrollment window and change rules plainly", "Compares tier costs and checks provider networks", "Tracks documents and deadlines so changes are not missed", "Prepares a draft election for the employee to submit" ], "outcomes": [ "Fewer missed windows: each employee sees their own deadline", "Clearer choices: tier costs and network gaps side by side", "Lighter help desk: consistent, sourced answers" ], "customer_scenario": [ "Employees miss the short window after a life event", "Comparing tier costs means per-paycheck math by hand", "Plans are chosen before doctors are checked in network", "Benefits staff field the same questions all season" ], "slide_title": "Benefits Enrollment Agent", "slide_summary": "Guides employees through open enrollment and life events, from change windows and plan costs to networks and a draft election they submit.", "scenario_name": "A mid-size employer", "challenge_intro": "Employees with a life event must change benefits quickly:", "actions_intro": "The Benefits Enrollment Agent:", "outcomes_intro": "Employees act on time with clearer choices:", "opportunity_statements": [ "Never miss a life-event change window", "Compare plan costs for your own coverage tier", "Check your doctors before you choose a plan", "Arrive at the benefits portal with a ready draft" ], "module_name": null, "module_description": null, "source_verified": true, "sales_headline": "Guide every employee through enrollment without making their benefits decisions for them", "card_pitch": "Give employees a conversational guide to the enrollment window, life-event rules, plan costs for their tier, provider networks, missing documents, and a draft election, while eligibility, plan choice, and submission stay with the employee and benefits staff.", "why_try": "The demo lets an employee ask a natural question and receive a source-backed, reviewable answer with the approval boundary visible in the response.", "customer_challenge_copy": "Employees miss life-event windows, compare plans by hand, and discover network gaps too late, while benefits staff answer the same questions all season.", "microsoft_ai_story": "Copilot Studio and Microsoft Teams give employees a conversational front door; approved benefits administration and carrier directory sources can be bound with identity, row-level access, data minimization, and audit controls.", "business_value": [ "Helps employees act within open enrollment and life-event windows.", "Makes plan costs and provider networks easy to compare for each employee.", "Reduces routine enrollment questions for benefits staff." ], "search_terms": [ "Benefits Enrollment Agent", "Benefits Specialist", "Employee", "HR Operations Staff", "Microsoft 365 Copilot", "Microsoft Copilot Studio", "Microsoft Teams", "benefits plans", "compare_plans", "deadline_reminders", "document_checklist", "election", "election_draft", "enrollment_window", "life event", "life_event_change", "open enrollment", "provider network", "provider_network" ], "journey_stage": "Empower", "blueprint_role": "Separates enrollment guidance and draft preparation from protected eligibility decisions and benefit elections.", "sample_prompts": [ { "label": "Employee: Enrollment window explanation", "prompt": "When does open enrollment close, and what can I change?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=benefits-enrollment-assistant-enrollment-window" }, { "label": "Employee: Life-event change rules", "prompt": "We just had a baby. What benefit changes can I make, and by when?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=benefits-enrollment-assistant-life-event-change" }, { "label": "Employee: Plan cost comparison", "prompt": "Compare the three medical plans for family coverage.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=benefits-enrollment-assistant-compare-plans" }, { "label": "Employee: Provider network check", "prompt": "Are Lakeside Family Clinic and Riverbend Pediatrics in network on each plan?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=benefits-enrollment-assistant-provider-network" }, { "label": "Employee: Document checklist", "prompt": "What documents do I still need to send?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=benefits-enrollment-assistant-document-checklist" }, { "label": "Employee: Election draft preview", "prompt": "Draft my election changes, but don't submit anything.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=benefits-enrollment-assistant-election-draft" }, { "label": "Employee: Personal deadline list", "prompt": "What benefits deadlines should I put on my calendar?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=benefits-enrollment-assistant-deadline-reminders" } ], "architecture": { "business_flow": [ "Employees and benefits specialists", "Microsoft Teams or Microsoft 365 Copilot", "Benefits Enrollment Agent", "Approved benefits administration system", "Carrier provider directories", "Employee submission and benefits staff review" ], "capabilities": [ { "operation": "enrollment_window", "name": "Enrollment window explanation", "purpose": "Use when an employee asks when open enrollment opens or closes, how many days are left, or what can change." }, { "operation": "life_event_change", "name": "Life-event change rules", "purpose": "Use when an employee states a life event such as a new child, marriage, loss of other coverage, or divorce and asks what can change and by when." }, { "operation": "compare_plans", "name": "Plan cost comparison", "purpose": "Use when an employee asks to compare medical plan costs, deductibles, out-of-pocket maximums, and networks for their coverage tier." }, { "operation": "provider_network", "name": "Provider network check", "purpose": "Use when an employee asks whether their doctors or clinics are in network on each plan." }, { "operation": "document_checklist", "name": "Document checklist", "purpose": "Use when an employee or benefits specialist asks which supporting documents are received, missing, and when they are due." }, { "operation": "election_draft", "name": "Election draft preview", "purpose": "Use when an employee asks to draft or preview election changes; the draft stays Not Submitted and shows the per-paycheck cost change." }, { "operation": "deadline_reminders", "name": "Personal deadline list", "purpose": "Use when an employee asks which benefits deadlines are coming up and in what order." } ], "easy_mode": [ "Verify Power Platform CLI 2.9.3 or newer and authenticate to the approved target environment.", "Initialize a modern cli-copilot project with an approved publisher prefix and preserve the generated identity settings.", "Upload the two packaged synthetic knowledge files and every operation-specific SKILL.md.", "Bind only approved least-privilege production connections after security, privacy, and business-owner review.", "Push the Draft agent, replay every locked persona case, and stop before publish for explicit approval." ], "local_install_prompt": "Install Benefits Enrollment Agent from https://raw.githubusercontent.com/microsoft/aibast-agents-library/main/solutions/benefits-enrollment-assistant/deployment.json, verify the expected tool and smoke case, and preserve the synthetic read-only boundary. Do not ask me to open a terminal. Do not connect customer systems or execute external actions.", "copilot_studio_prompt": "Using the Microsoft Copilot Studio plugin, initialize a Draft agent for Benefits Enrollment Agent from https://raw.githubusercontent.com/microsoft/aibast-agents-library/main/solutions/benefits-enrollment-assistant/deployment.json. Upload the two synthetic knowledge files and 7 skills, replay every locked case, and Stop before publish. Do not bind production connections without explicit customer approval.", "required_connections": [ "Use only approved, least-privilege production connections; the packaged demonstration uses synthetic knowledge files.", "Approved benefits administration system (read-only, identity-scoped)", "Approved carrier provider directory (read-only)" ], "manual_commands": [ "pac auth create", "pac copilot init --name \"Benefits Enrollment Agent\" --publisher-prefix --authoring-mode cli-copilot --project-dir \"\" --environment \"\"", "pac connection list --environment \"\"", "pac copilot pull --project-dir \"\"", "pac copilot push --project-dir \"\"", "pac copilot publish --bot \"\" --environment \"\"" ], "acceptance_checks": [ "Every exposed operation has one persona-language case and one SKILL.md.", "Exactly two synthetic knowledge files are present.", "Responses state the relevant human approval or review boundary.", "No operation claims an external write, communication, commitment, or protected decision.", "Publishing remains a separate user-approved step." ], "hard_mode": [ "Run pac copilot init and verify settings.mcs.yml before authoring components.", "Author global instructions that preserve the packaged synthetic-only and human-approval boundaries.", "Add the two knowledge files and one skill for every exposed operation.", "Create dedicated connection references only for approved systems and verify least-privilege access.", "Pull before push, inspect every generated file, replay every locked case, and stop before publish." ] }, "curated_copy": true, "source": "field", "origin": "both", "onepager": "63. Benefits Enrollment Agent one-pager.pptx", "demo_video": "63-Benefits Enrollment Agent.mp4", "has_onepager": true, "has_demo_video": true, "promise_coverage": 100, "is_primary": true, "aliases": [], "package": { "slug": "benefits-enrollment-assistant", "quest_url": "solutions/benefits-enrollment-assistant/quest.html", "manual_tutorial_url": "solutions/benefits-enrollment-assistant/manual-tutorial.html", "export_manifest_url": "solutions/benefits-enrollment-assistant/export-manifest.json" } }, "_demo": { "slug": "benefits-enrollment-assistant", "case_count": 7, "personas": [ "Employee" ], "prompts": [ "When does open enrollment close, and what can I change?", "We just had a baby. What benefit changes can I make, and by when?", "Compare the three medical plans for family coverage." ] }, "_catalog_kind": "solution", "_readiness": [ "Demo proven", "Conversation tested", "Synthetic data", "Production pattern" ] }, { "schema": "rapp-agent/1.0", "name": "@aibast-agents-library/build-materials-compliance", "version": "1.0.0", "display_name": "Build Materials Compliance Agent", "description": "Keep grant-funded construction bills of materials compliant every week: flag items that newly fall out of the program's materials sourcing rule with cited evidence, find same-specification compliant alternates, watch vendor certificates, and keep the audit evidence package ready.", "author": "AIBAST", "tags": [ "manufacturing", "construction", "materials", "compliance", "procurement", "bill-of-materials" ], "category": "manufacturing", "quality_tier": "verified", "requires_env": [], "dependencies": [ "@rapp/basic-agent" ], "_file": "agents/@aibast-agents-library/manufacturing_stacks/build_materials_compliance_stack/build_materials_compliance_agent.py", "_size_kb": 22.7, "_lines": 411, "_sha256": "974728e68dd164dbf119b3a5c9a6179e085024561f1a28496308bd616ecd113c", "_install_prefix": "build_materials_compliance__", "_install_filename": "build_materials_compliance__974728e68dd1_agent.py", "_stack": "build_materials_compliance", "_stack_vertical": "manufacturing", "_synthetic_data": true, "_added_at": "2026-10-08T10:30:56-04:00", "_solution": { "advertised_name": "Build Materials Compliance Agent", "sharepoint_list_name": "Build Materials Compliance Agent", "slot": 53, "executive_summary": "Keep grant-funded construction bills of materials compliant every week: flag items that newly fall out of the program's materials sourcing rule with cited evidence, find same-specification compliant alternates, watch vendor certificates, and keep the audit evidence package ready.", "industries": [ "Manufacturing", "Construction", "Public Infrastructure" ], "personas": [ "Procurement Compliance Leads", "Buyers", "Supplier Quality Managers", "Program Compliance Reviewers" ], "featured_tools": [ "Microsoft Teams", "SharePoint", "Microsoft Word" ], "agent_requirements": [ "Microsoft Copilot Studio" ], "capabilities": [ "Scans approved bills of materials and reports what changed", "Flags newly non-compliant items with the cited evidence", "Finds same-spec compliant alternates and sizes the switch", "Keeps certificates and audit evidence current and drafts approved next steps" ], "outcomes": [ "Catches compliance problems before materials are ordered", "Ties every finding to its evidence", "Stays audit-ready every week" ], "customer_scenario": [ "Weekly BOM changes are hard to track across projects", "A lapsed certificate is found after the order is placed", "Finding a compliant same-spec alternative takes days", "Evidence packages have gaps nobody sees until the audit" ], "slide_title": "Build Materials Compliance Agent", "slide_summary": "Keeps grant-funded construction bills of materials compliant weekly, with cited findings, alternates and audit-ready evidence.", "scenario_name": "An infrastructure contractor", "challenge_intro": "The procurement team checks compliance by hand:", "actions_intro": "The Build Materials Compliance Agent:", "outcomes_intro": "With the agent the team:", "opportunity_statements": [ "Covered materials must meet the program's sourcing rule", "Vendor certificates lapse and answers change without notice", "Bills of materials change every week", "Audit evidence is rebuilt by hand at audit time" ], "module_name": null, "module_description": null, "source_verified": true, "sales_headline": "Catch materials compliance problems the week they appear", "card_pitch": "Give a contractor's procurement team one assistant that scans the approved bills of materials every week, flags items that newly fall out of the program's materials sourcing rule with cited evidence, finds same-spec compliant alternates, and keeps the audit evidence package ready.", "why_try": "The demo runs a full weekly compliance cycle in six plain questions, from what changed to approved drafts, and every finding names the record behind it.", "customer_challenge_copy": "Grant-funded projects must prove that covered materials meet a sourcing rule, but certificates lapse, vendor answers change and bills of materials move every week, so problems surface at audit time instead of before the order.", "microsoft_ai_story": "Microsoft Copilot Studio hosts the conversation in Microsoft Teams; approved connections to SharePoint document libraries for certificates and evidence, product master data and Word drafts can be bound later with least-privilege access and human approval.", "business_value": [ "Finds newly non-compliant materials in the week they change, before they are ordered.", "Turns a lapse into a ready same-spec alternative with its cost impact sized.", "Keeps the audit evidence package current instead of rebuilding it at audit time." ], "search_terms": [ "Build Materials Compliance Agent", "Buyer", "Microsoft Copilot Studio", "Microsoft Teams", "Microsoft Word", "Procurement Compliance Lead", "Program Compliance Reviewer", "SharePoint", "Supplier Quality Manager", "action_drafts", "audit evidence", "audit_evidence", "bill of materials", "compliance_check", "compliant_alternates", "construction", "materials compliance", "procurement", "vendor certificate", "vendor_watch", "weekly_changes" ], "journey_stage": "Optimize", "blueprint_role": "Establishes a governed weekly compliance loop that separates scanning, evidence and recommendations from ordering, master-data writes and compliance determinations.", "sample_prompts": [ { "label": "Procurement Compliance Lead: Weekly BOM change scan", "prompt": "What changed on our approved bills of materials this week?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=build-materials-compliance-weekly-changes" }, { "label": "Procurement Compliance Lead: Materials compliance check", "prompt": "Is anything newly out of compliance, and why?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=build-materials-compliance-compliance-check" }, { "label": "Buyer: Compliant alternates", "prompt": "Find me a compliant replacement for the gate valve with the same spec.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=build-materials-compliance-compliant-alternates" }, { "label": "Supplier Quality Manager: Vendor certificate watch", "prompt": "Which vendor certificates have lapsed or are about to?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=build-materials-compliance-vendor-watch" }, { "label": "Program Compliance Reviewer: Audit evidence readiness", "prompt": "Are we ready if the program auditor asks for our evidence package?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=build-materials-compliance-audit-evidence" }, { "label": "Procurement Compliance Lead: Approved next-step drafts", "prompt": "I approve the hold, the GV-08B substitution and the vendor follow-ups. Prepare the drafts.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=build-materials-compliance-action-drafts" } ], "architecture": { "business_flow": [ "Approved bills of materials", "Microsoft Teams or Microsoft 365 Copilot", "Build Materials Compliance Agent", "Vendor certificates and questionnaires", "Recommendations and drafts for approval", "Procurement and program reviewers" ], "capabilities": [ { "operation": "weekly_changes", "name": "Weekly BOM change scan", "purpose": "Use when someone asks what is new or changed on the approved bills of materials this week." }, { "operation": "compliance_check", "name": "Materials compliance check", "purpose": "Use when someone asks which items are out of compliance or newly non-compliant and why." }, { "operation": "compliant_alternates", "name": "Compliant alternates", "purpose": "Use when someone asks for a compliant replacement or alternate part with the same specification." }, { "operation": "vendor_watch", "name": "Vendor certificate watch", "purpose": "Use when someone asks which vendor certificates have expired or expire soon." }, { "operation": "audit_evidence", "name": "Audit evidence readiness", "purpose": "Use when someone asks whether the audit evidence package is ready for each project." }, { "operation": "action_drafts", "name": "Approved next-step drafts", "purpose": "Use when a person has approved next steps and wants the hold notice, renewal, substitution and document request drafts." } ], "easy_mode": [ "Verify Power Platform CLI 2.9.3 or newer and authenticate to the approved target environment.", "Initialize a modern cli-copilot project with an approved publisher prefix and preserve the generated identity settings.", "Upload the two packaged synthetic knowledge files and every operation-specific SKILL.md.", "Bind only approved least-privilege production connections after security, privacy, and business-owner review.", "Push the Draft agent, replay every locked persona case, and stop before publish for explicit approval." ], "local_install_prompt": "Install Build Materials Compliance Agent from https://raw.githubusercontent.com/microsoft/aibast-agents-library/main/solutions/build-materials-compliance/deployment.json, verify the expected tool and smoke case, and preserve the synthetic read-only boundary. Do not ask me to open a terminal. Do not connect customer systems or execute external actions.", "copilot_studio_prompt": "Using the Microsoft Copilot Studio plugin, initialize a Draft agent for Build Materials Compliance Agent from https://raw.githubusercontent.com/microsoft/aibast-agents-library/main/solutions/build-materials-compliance/deployment.json. Upload the two synthetic knowledge files and 6 skills, replay every locked case, and Stop before publish. Do not bind production connections without explicit customer approval.", "required_connections": [ "SharePoint document library for certificates and evidence (read only)", "Product master data (write after human approval)", "Word drafts in an action folder" ], "manual_commands": [ "pac auth create", "pac copilot init --name \"Build Materials Compliance Agent\" --publisher-prefix --authoring-mode cli-copilot --project-dir \"\" --environment \"\"", "pac connection list --environment \"\"", "pac copilot pull --project-dir \"\"", "pac copilot push --project-dir \"\"", "pac copilot publish --bot \"\" --environment \"\"" ], "acceptance_checks": [ "Every exposed operation has one persona-language case and one SKILL.md.", "Exactly two synthetic knowledge files are present.", "Responses state the relevant human approval or review boundary.", "No operation claims an external write, communication, commitment, or protected decision.", "Publishing remains a separate user-approved step." ], "hard_mode": [ "Run pac copilot init and verify settings.mcs.yml before authoring components.", "Author global instructions that preserve the packaged synthetic-only and human-approval boundaries.", "Add the two knowledge files and one skill for every exposed operation.", "Create dedicated connection references only for approved systems and verify least-privilege access.", "Pull before push, inspect every generated file, replay every locked case, and stop before publish." ] }, "curated_copy": true, "source": "field", "origin": "both", "onepager": "53. Build Materials Compliance Agent one-pager.pptx", "demo_video": "53-Build Materials Compliance Agent.mp4", "has_onepager": true, "has_demo_video": true, "promise_coverage": 100, "is_primary": true, "aliases": [], "package": { "slug": "build-materials-compliance", "quest_url": "solutions/build-materials-compliance/quest.html", "manual_tutorial_url": "solutions/build-materials-compliance/manual-tutorial.html", "export_manifest_url": "solutions/build-materials-compliance/export-manifest.json" } }, "_demo": { "slug": "build-materials-compliance", "case_count": 6, "personas": [ "Buyer", "Procurement Compliance Lead", "Program Compliance Reviewer", "Supplier Quality Manager" ], "prompts": [ "What changed on our approved bills of materials this week?", "Is anything newly out of compliance, and why?", "Find me a compliant replacement for the gate valve with the same spec." ] }, "_catalog_kind": "solution", "_readiness": [ "Demo proven", "Conversation tested", "Synthetic data", "Production pattern" ] }, { "schema": "rapp-agent/1.0", "name": "@aibast-agents-library/inventory-rebalancing", "version": "1.0.0", "display_name": "Inventory Rebalancing Agent", "description": "Analyze a fixed synthetic inventory snapshot and recommend review-ready rebalancing options. Never claim that inventory was moved, reordered, returned, or liquidated.", "author": "AIBAST", "tags": [ "inventory", "warehouse", "supply-chain", "rebalancing", "manufacturing" ], "category": "manufacturing", "quality_tier": "verified", "requires_env": [], "dependencies": [ "@rapp/basic-agent" ], "_file": "agents/@aibast-agents-library/manufacturing_stacks/inventory_rebalancing_stack/inventory_rebalancing_agent.py", "_size_kb": 32.2, "_lines": 662, "_sha256": "0315b2989a652f9c3255bb45815b0ec3bc95729bde2adf73f1165dd17c2ea89d", "_install_prefix": "inventory_rebalancing__", "_install_filename": "inventory_rebalancing__0315b2989a65_agent.py", "_stack": "inventory_rebalancing", "_stack_vertical": "manufacturing", "_synthetic_data": true, "_added_at": "2026-03-18T13:23:12-04:00", "_solution": { "advertised_name": "Inventory Rebalancing Agent", "sharepoint_list_name": "Inventory Optimization Agent", "slot": 7, "executive_summary": "Intelligently optimize inventory portfolios to improve cash flow and warehouse efficiency while reducing waste.", "industries": [ "Manufacturing" ], "personas": [ "Supply Chain Managers", "Inventory Managers", "Procurement Manager" ], "featured_tools": [ "Dynamics 365", "Microsoft Teams" ], "agent_requirements": [ "Microsoft Copilot Studio" ], "capabilities": [ "Analyze inventory and classify items by velocity and strategic value", "Develop a phased recovery plan with flash sales, vendor returns, and safety stock adjustments", "Optimize reorder schedules", "Implement dynamic inventory policies and AI-driven alerts" ], "outcomes": [ "Freed up working capital", "Improved warehouse efficiency with space optimization and streamlined picking processes", "Reduced financial risk by minimizing obsolescence", "Enhanced operational agility with dynamic inventory policies and AI-driven demand alignment" ], "customer_scenario": [ "An inventory portfolio burdened with slow-moving and obsolete stock", "Warehouse operating at near-critical capacity that caused congestion", "Rising storage costs and increasing obsolescence risk" ], "slide_title": "Inventory Rebalancing Agent", "slide_summary": "Intelligently optimize inventory portfolios to improve cash flow and warehouse efficiency while reducing waste.", "scenario_name": "A consumer goods manufacturer", "challenge_intro": "A manufacturer needed a smarter way to identify and clear out excess inventory. They ran into multiple issues:", "actions_intro": "The manufacturer used the Inventory Rebalancing Agent to:", "outcomes_intro": "With the Agent, the firm:", "opportunity_statements": [ "Identify and classify slow-moving, obsolete, and excess inventory to quantify where working capital is tied up", "Design liquidation and return strategies to unlock cash and reduce waste for healthier financial performance", "Apply AI-driven policies and dynamic reorder points to maintain optimal inventory levels and efficiency gains" ], "module_name": null, "module_description": null, "source_verified": true, "sales_headline": "Find inventory pressure before it becomes a service or capacity problem", "card_pitch": "Turn warehouse capacity, SKU position, demand, lifecycle, and cost evidence into review-ready balancing options without moving stock or changing inventory policy.", "why_try": "It makes inventory AI concrete: supply chain teams can ask where pressure is building, see the relevant SKU evidence, and compare transfer or policy-review options with explicit approval gates.", "customer_challenge_copy": "A manufacturer lacks one shared view of warehouse pressure, forecast-relative imbalance, slow-moving stock, lifecycle risk, and planning cost. Manual reconciliation delays decisions and can blur the line between analysis and authorized inventory action.", "microsoft_ai_story": "Microsoft Copilot Studio provides the governed conversational experience. Dynamics 365 Supply Chain Management contributes inventory, demand, and policy context; warehouse and transportation systems support approved execution; Power BI presents portfolio and capacity views; Microsoft Teams supports review and approval.", "business_value": [ "Improves visibility into facility pressure, inventory imbalance, and lifecycle signals that deserve planning attention.", "Accelerates cross-functional review of transfer, reorder, return, and disposition options using consistent evidence.", "Strengthens inventory governance by separating recommendations from reservations, movements, orders, and policy changes." ], "search_terms": [ "inventory rebalancing", "inventory manager", "supply chain manager", "warehouse capacity", "slow-moving inventory", "excess inventory", "inventory shortage", "inventory transfer planning", "reorder policy", "Dynamics 365 Supply Chain Management", "Power BI inventory", "Copilot Studio manufacturing" ], "journey_stage": "Optimize", "blueprint_role": "Creates a governed inventory intelligence layer that connects warehouse, demand, lifecycle, and cost evidence while preserving ERP and warehouse-management authorization boundaries.", "sample_prompts": [ { "label": "Review warehouse and SKU pressure", "prompt": "Which distribution centers are tight on space, and which SKU positions should my team review first?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=inventory-capacity-review" }, { "label": "Find shortages and slow-moving stock", "prompt": "Where do we have forecast-relative shortages or excess that deserve a rebalancing review?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=inventory-imbalance-review" }, { "label": "Prepare proposed warehouse moves", "prompt": "Show me the proposed warehouse moves, but do not move or reserve anything.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=inventory-transfer-plan" }, { "label": "Compare inventory planning trade-offs", "prompt": "Where is inventory exposure concentrated, and what trade-offs should I take to the planning meeting?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=inventory-cost-tradeoffs" } ], "architecture": { "business_flow": [ "Supply chain, inventory, procurement, and warehouse leaders", "Microsoft 365 Copilot or Copilot Studio", "Inventory Rebalancing Agent", "Dynamics 365 Supply Chain Management", "Warehouse and transportation management systems", "Power BI", "Authorized inventory and finance reviewers" ], "capabilities": [ { "operation": "inventory_snapshot", "name": "Inventory and capacity overview", "purpose": "Summarizes facility utilization, SKU levels, reorder exposure, and inventory value from approved records." }, { "operation": "rebalance_recommendation", "name": "Portfolio classification and imbalance review", "purpose": "Explains forecast-relative shortages and excess alongside synthetic velocity, strategic-value, and lifecycle signals." }, { "operation": "transfer_plan", "name": "Proposed transfer planning", "purpose": "Prepares review-ready inter-warehouse movement options without reserving, picking, shipping, or moving inventory." }, { "operation": "cost_analysis", "name": "Inventory planning trade-off analysis", "purpose": "Compares synthetic holding, shortage, and transfer estimates while avoiding customer-outcome commitments." } ], "easy_mode": [ "Verify Power Platform CLI 2.9.3 or newer and authenticate to the target environment.", "Use the Copilot Studio plugin initialization workflow to create a modern, sync-connected agent project with an approved publisher prefix.", "Give the Copilot Studio Architect this blueprint and the initialized project so it writes governed instructions, operation-specific skills, and approval-aware tool definitions.", "Bind approved inventory, demand, warehouse, transportation, reporting, and collaboration connections through dedicated connection references.", "Use the Copilot Studio Manage workflow to push the project, replay every locked prompt, confirm that no inventory side effect is claimed, and stop before publishing." ], "local_install_prompt": "You are GitHub Copilot running in Agent mode in VS Code. Install and open the AI BAST Inventory Rebalancing Agent from this deployment recipe: https://raw.githubusercontent.com/microsoft/aibast-agents-library/main/solutions/inventory-rebalancing/deployment.json. Own the entire setup. Do not ask me to open a terminal, run a command, clone a repository, or install the Brainstem myself. Inspect the operating system; check the recipe health endpoint; if the Brainstem is missing, run the official installer yourself using the recipe's OS-appropriate installer; install or refresh the policy-clean launcher; fetch registry.json and the raw agent source; verify the file against the registry SHA-256; place it under the discovered Brainstem agents directory with the recipe target filename; start the Brainstem; verify /health reports the expected tool; run the recipe smoke prompt through /chat and confirm the expected agent fired; then open the local chat UI in my browser. If authentication requires a browser confirmation, launch the supported device flow and ask me only to complete that browser step. Do not publish, connect customer systems, move, reserve, reorder, return, liquidate, or otherwise change inventory, or silently substitute another agent. If any step fails, stop and show the exact failed step and output.", "copilot_studio_prompt": "You are GitHub Copilot running in Agent mode in VS Code with the Microsoft Copilot Studio plugin installed. Promote the AI BAST Inventory Rebalancing Agent using this deployment recipe: https://raw.githubusercontent.com/microsoft/aibast-agents-library/main/solutions/inventory-rebalancing/deployment.json. Own every command and file change; do not ask me to use a terminal. Ask me only for the target Power Platform environment ID and publisher prefix. Verify PAC CLI is at least 2.9.3, authenticate if needed, initialize a new cli-copilot project, and use the Copilot Studio Architect workflow to author modern YAML that preserves inventory_snapshot, rebalance_recommendation, transfer_plan, and cost_analysis. Bind approved Dynamics 365 Supply Chain Management, warehouse, transportation, Power BI, and Microsoft Teams connections. Review every generated YAML file and connection, pull before push, run pac copilot push, and replay every locked prompt. Confirm that all transfer, reorder, return, disposition, and policy outputs remain proposals requiring authorization. Stop before publish and show the generated files, validation evidence, unresolved connection work, and the exact publish command that would make it live.", "required_connections": [ "Dynamics 365 Supply Chain Management or another approved ERP inventory source", "Approved warehouse and transportation management systems", "Power BI", "Microsoft Teams" ], "manual_commands": [ "pac auth create", "pac copilot init --name \"Inventory Rebalancing Agent\" --publisher-prefix --authoring-mode cli-copilot --project-dir \"\" --environment \"\"", "pac connection list --environment \"\"", "pac copilot pull --project-dir \"\"", "pac copilot push --project-dir \"\"", "pac copilot publish --bot \"\" --environment \"\"" ], "acceptance_checks": [ "settings.mcs.yml exists and preserves the initialized identity, environment binding, language, and authoring mode.", "Every implemented inventory operation is represented by instructions, an uploadable skill, or a bound tool.", "Both synthetic knowledge files are loaded and clearly label every exact figure and identifier as fictional pilot evidence.", "Inventory, demand, warehouse, transportation, reporting, and collaboration sources use dedicated and verified connection references.", "The capacity, imbalance, transfer-plan, and cost-trade-off prompts return the expected synthetic records.", "No output claims that inventory was reserved, moved, reordered, returned, liquidated, or governed by a changed policy.", "Public value statements remain qualitative and publishing remains a separate, user-approved step." ], "hard_mode": [ "Run pac copilot init and confirm settings.mcs.yml exists.", "Author global instructions that separate inventory evidence, classification, recommendation, authorization, and external execution.", "Upload the two synthetic Markdown knowledge files and every operation-specific SKILL.md file from this package.", "Create read tools for approved inventory, demand, warehouse, transportation, and reporting data with dedicated connection references.", "Implement explicit approval and action tools for any future reservation, movement, reorder, return, disposition, or policy change; keep them unavailable in the recommendation-only pilot.", "Validate every component schema and filename, run pac copilot push, inspect the generated agent, replay all locked prompts, and stop before publish." ] }, "curated_copy": true, "source": "catalog", "origin": null, "onepager": "Inventory Rebalancing Agent one-pager.pptx", "demo_video": "#7-Inventory Rebalancing Agent.mp4", "has_onepager": true, "has_demo_video": true, "promise_coverage": 55, "is_primary": true, "aliases": [], "package": { "slug": "inventory-rebalancing", "quest_url": "solutions/inventory-rebalancing/quest.html", "manual_tutorial_url": "solutions/inventory-rebalancing/manual-tutorial.html", "export_manifest_url": "solutions/inventory-rebalancing/export-manifest.json" } }, "_demo": { "slug": "inventory-rebalancing", "case_count": 10, "personas": [ "Inventory Manager", "Procurement Manager", "Supply Chain Manager" ], "prompts": [ "Which distribution centers are tight on space, and which SKU positions should my team review first?", "Where do we have forecast-relative shortages or excess that deserve a rebalancing review?", "Show me the proposed warehouse moves, but do not move or reserve anything." ] }, "_catalog_kind": "solution", "_readiness": [ "Demo proven", "Conversation tested", "Synthetic data", "Production pattern" ] }, { "schema": "rapp-agent/1.0", "name": "@aibast-agents-library/maintenance-scheduling", "version": "1.0.0", "display_name": "Maintenance Scheduling Agent", "description": "Analyze a fixed synthetic maintenance snapshot and prepare review-ready alerts, schedule options, cost cases, draft work orders, calendars and optimization plans. Never control equipment, create or dispatch work orders, assign technicians, send notifications, or reserve parts.", "author": "AIBAST", "tags": [ "maintenance", "predictive", "scheduling", "manufacturing", "IoT" ], "category": "manufacturing", "quality_tier": "verified", "requires_env": [], "dependencies": [ "@rapp/basic-agent" ], "_file": "agents/@aibast-agents-library/manufacturing_stacks/maintenance_scheduling_stack/maintenance_scheduling_agent.py", "_size_kb": 26.9, "_lines": 536, "_sha256": "f364d49b970b319fc9d0de5e25fc95a680008ef77b36a6782fd8d6d7b737d510", "_install_prefix": "maintenance_scheduling__", "_install_filename": "maintenance_scheduling__f364d49b970b_agent.py", "_stack": "maintenance_scheduling", "_stack_vertical": "manufacturing", "_synthetic_data": true, "_added_at": "2026-03-18T13:23:12-04:00", "_solution": { "advertised_name": "Maintenance Scheduling Agent", "sharepoint_list_name": "Predictive Maintenance Agent", "slot": 15, "executive_summary": "Perform predictive maintenance analysis and scheduling orchestration to prevent unplanned downtime and protect production capacity.", "industries": [ "Manufacturing" ], "personas": [ "Maintenance Manager", "Production Supervisor", "Operation Leader" ], "featured_tools": [ "Dynamics 365", "Microsoft Teams" ], "agent_requirements": [ "Microsoft Copilot Studio" ], "capabilities": [ "Analyze IoT sensor data and predict time-to-failure", "Identify optimal maintenance windows", "Automate crew assignments and parts staging for readiness", "Validate backup equipment availability" ], "outcomes": [ "Prevented unplanned downtime and maintained delivery commitments", "Reduced emergency repair costs through proactive scheduling", "Extended equipment lifespan while sustaining quality standards", "Improved production schedule adherence, reducing disruptions" ], "customer_scenario": [ "Wear indicators signaled imminent equipment failure", "Manual scheduling risked disrupting delivery timelines", "Parts availability and crew coordination were unclear", "Maintenance planning lacked full visibility into capacity constraints" ], "slide_title": "Maintenance Scheduling Agent", "slide_summary": "Perform predictive maintenance analysis and scheduling orchestration to prevent unplanned downtime and protect production capacity.", "scenario_name": "A global auto manufacturer", "challenge_intro": "An auto manufacturer had to prevent downtime during a key production run:", "actions_intro": "The manufacturer used the Maintenance Scheduling Agent to:", "outcomes_intro": "With the agent\u2019s help, the plant:", "opportunity_statements": [ "Identify predictive maintenance windows to minimize production impact and extend equipment life", "Coordinate resources and parts staging to ensure timely execution and reduce emergency repair costs", "Deliver real-time alerts and scheduling insights to prevent disruptions and maintain quality standards" ], "module_name": null, "module_description": null, "source_verified": true, "sales_headline": "Bring condition, capacity, and readiness evidence into one maintenance decision", "card_pitch": "Turn equipment condition, failure-risk, technician capacity, parts readiness, backup capacity, and downtime evidence into review-ready maintenance options without controlling equipment or dispatching work.", "why_try": "It makes predictive-maintenance AI concrete: maintenance and production leaders can identify the assets needing attention and review crew, parts, backup, and production-impact considerations together.", "customer_challenge_copy": "A manufacturer reviews condition signals, maintenance history, technician capacity, parts readiness, backup equipment, and production constraints in separate systems. Manual coordination slows planning and can imply assignments or dispatch before authorized review.", "microsoft_ai_story": "Microsoft Copilot Studio provides the governed conversational experience. Azure IoT Hub contributes approved condition signals, Dynamics 365 Supply Chain Management and Field Service contribute asset, parts, and work context, Power BI presents risk and downtime views, and Microsoft Teams supports planner and production approvals.", "business_value": [ "Improves visibility into equipment condition and the operational context surrounding maintenance risk.", "Accelerates coordinated review of maintenance windows, technician candidates, parts, and backup capacity.", "Protects operational control by keeping equipment commands, work creation, assignment, scheduling, and dispatch in approved systems." ], "search_terms": [ "maintenance scheduling", "predictive maintenance", "maintenance manager", "production supervisor", "equipment condition", "failure risk", "work order planning", "technician capacity", "parts readiness", "Azure IoT Hub", "Dynamics 365 Field Service", "Copilot Studio manufacturing" ], "journey_stage": "Optimize", "blueprint_role": "Creates a governed maintenance intelligence layer that joins equipment signals with work, resource, parts, backup, and production context while preserving CMMS, MES, and equipment-control boundaries.", "sample_prompts": [ { "label": "Review equipment and technician capacity", "prompt": "Give me the equipment and technician-capacity picture I should review before the weekly maintenance meeting.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=maintenance-capacity-review" }, { "label": "Find asset signals needing review", "prompt": "Which asset signals deserve immediate human review before the next production run?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=maintenance-predictive-alerts" }, { "label": "Draft maintenance candidates safely", "prompt": "Draft the maintenance candidates with crew, parts, and backup checks, but do not create or dispatch work orders.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=maintenance-work-plan" }, { "label": "Compare downtime trade-offs", "prompt": "What synthetic downtime exposure and preventive trade-offs should leadership review?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=maintenance-downtime-analysis" } ], "architecture": { "business_flow": [ "Maintenance managers, production supervisors, and operations leaders", "Microsoft 365 Copilot or Copilot Studio", "Maintenance Scheduling Agent", "Azure IoT Hub and approved condition monitoring", "Dynamics 365 asset, parts, and work management", "MES and production schedules", "Authorized maintenance and production reviewers" ], "capabilities": [ { "operation": "schedule_overview", "name": "Maintenance and resource overview", "purpose": "Summarizes equipment status, risk, maintenance history, and technician capacity for planning review." }, { "operation": "predictive_alerts", "name": "Predictive condition alerts", "purpose": "Ranks synthetic sensor and failure-risk evidence while requiring validation in approved condition-monitoring systems." }, { "operation": "work_order_plan", "name": "Proposed maintenance work planning", "purpose": "Frames maintenance candidates, technician candidates, parts checks, and backup checks without creating, assigning, scheduling, or dispatching work." }, { "operation": "downtime_analysis", "name": "Downtime and preventive trade-off analysis", "purpose": "Compares synthetic exposure and preventive estimates without promising avoided downtime, savings, or production outcomes." } ], "easy_mode": [ "Verify Power Platform CLI 2.9.3 or newer and authenticate to the target environment.", "Use the Copilot Studio plugin initialization workflow to create a modern, sync-connected agent project with an approved publisher prefix.", "Give the Copilot Studio Architect this blueprint and the initialized project so it writes governed instructions, operation-specific skills, and approval-aware tool definitions.", "Bind approved condition-monitoring, asset, parts, work-management, production-schedule, reporting, and collaboration connections through dedicated connection references.", "Use the Copilot Studio Manage workflow to push the project, replay every locked prompt, confirm that no equipment control or dispatch is claimed, and stop before publishing." ], "local_install_prompt": "You are GitHub Copilot running in Agent mode in VS Code. Install and open the AI BAST Maintenance Scheduling Agent from this deployment recipe: https://raw.githubusercontent.com/microsoft/aibast-agents-library/main/solutions/maintenance-scheduling/deployment.json. Own the entire setup. Do not ask me to open a terminal, run a command, clone a repository, or install the Brainstem myself. Inspect the operating system; check the recipe health endpoint; if the Brainstem is missing, run the official installer yourself using the recipe's OS-appropriate installer; install or refresh the policy-clean launcher; fetch registry.json and the raw agent source; verify the file against the registry SHA-256; place it under the discovered Brainstem agents directory with the recipe target filename; start the Brainstem; verify /health reports the expected tool; run the recipe smoke prompt through /chat and confirm the expected agent fired; then open the local chat UI in my browser. If authentication requires a browser confirmation, launch the supported device flow and ask me only to complete that browser step. Do not publish, connect customer systems, control equipment, create or dispatch work orders, assign technicians, reserve parts, or alter production schedules, or silently substitute another agent. If any step fails, stop and show the exact failed step and output.", "copilot_studio_prompt": "You are GitHub Copilot running in Agent mode in VS Code with the Microsoft Copilot Studio plugin installed. Promote the AI BAST Maintenance Scheduling Agent using this deployment recipe: https://raw.githubusercontent.com/microsoft/aibast-agents-library/main/solutions/maintenance-scheduling/deployment.json. Own every command and file change; do not ask me to use a terminal. Ask me only for the target Power Platform environment ID and publisher prefix. Verify PAC CLI is at least 2.9.3, authenticate if needed, initialize a new cli-copilot project, and use the Copilot Studio Architect workflow to author modern YAML that preserves schedule_overview, predictive_alerts, work_order_plan, and downtime_analysis. Bind approved Azure IoT Hub, condition-monitoring, Dynamics 365 asset and work-management, parts, MES, Power BI, and Microsoft Teams connections. Review every generated YAML file and connection, pull before push, run pac copilot push, and replay every locked prompt. Confirm that equipment control, work creation, assignment, scheduling, dispatch, parts reservation, and production changes remain unavailable without explicit authorized tools. Stop before publish and show the generated files, validation evidence, unresolved connection work, and the exact publish command that would make it live.", "required_connections": [ "Azure IoT Hub or another approved condition-monitoring source", "Dynamics 365 asset, parts, and work-management data", "Approved MES and production schedules", "Power BI", "Microsoft Teams" ], "manual_commands": [ "pac auth create", "pac copilot init --name \"Maintenance Scheduling Agent\" --publisher-prefix --authoring-mode cli-copilot --project-dir \"\" --environment \"\"", "pac connection list --environment \"\"", "pac copilot pull --project-dir \"\"", "pac copilot push --project-dir \"\"", "pac copilot publish --bot \"\" --environment \"\"" ], "acceptance_checks": [ "settings.mcs.yml exists and preserves the initialized identity, environment binding, language, and authoring mode.", "Every implemented maintenance operation is represented by instructions, an uploadable skill, or a bound tool.", "Both synthetic knowledge files are loaded and clearly label every exact figure and identifier as fictional pilot evidence.", "Condition, asset, work, parts, production, reporting, and collaboration sources use dedicated and verified connection references.", "The overview, alert, proposed-work, and downtime prompts return the expected synthetic records.", "No output claims equipment control, work-order creation, assignment, scheduling, dispatch, parts reservation, or production-schedule change.", "Public value statements remain qualitative and publishing remains a separate, user-approved step." ], "hard_mode": [ "Run pac copilot init and confirm settings.mcs.yml exists.", "Author global instructions that separate condition evidence, maintenance recommendations, planner approval, safety review, and external execution.", "Upload the two synthetic Markdown knowledge files and every operation-specific SKILL.md file from this package.", "Create read tools for approved condition, asset, work, parts, technician, backup, and production data with dedicated connection references.", "Implement explicit authorization, safety, and action tools for any future equipment command, work creation, assignment, scheduling, dispatch, reservation, or production change; keep them unavailable in the recommendation-only pilot.", "Validate every component schema and filename, run pac copilot push, inspect the generated agent, replay all locked prompts, and stop before publish." ] }, "curated_copy": true, "source": "catalog", "origin": null, "onepager": "Maintenance Scheduling Agent one-pager.pptx", "demo_video": "18-Maintenance Scheduling Agent.mp4", "has_onepager": true, "has_demo_video": true, "promise_coverage": 60, "is_primary": true, "aliases": [], "package": { "slug": "maintenance-scheduling", "quest_url": "solutions/maintenance-scheduling/quest.html", "manual_tutorial_url": "solutions/maintenance-scheduling/manual-tutorial.html", "export_manifest_url": "solutions/maintenance-scheduling/export-manifest.json" } }, "_demo": { "slug": "maintenance-scheduling", "case_count": 7, "personas": [ "Maintenance Manager", "Operations Leader", "Production Supervisor" ], "prompts": [ "Give me the equipment and technician-capacity picture I should review before the weekly maintenance meeting.", "Machine #7 on Line 3 is showing wear indicators and we have a big order next week. What does the condition data say?", "Show me the lowest-impact maintenance window for machine #7 with coverage, crew, parts and backup, but do not create or dispatch work orders." ] }, "_catalog_kind": "solution", "_readiness": [ "Demo proven", "Conversation tested", "Synthetic data", "Production pattern" ] }, { "schema": "rapp-agent/1.0", "name": "@aibast-agents-library/order-entry-validation", "version": "1.0.0", "display_name": "Order Entry Validation Agent", "description": "Turn inbound customer purchase orders into validated draft sales orders: extract every PO field, check prices and terms against the accepted quote, catch invalid product configurations, classify the order, and hand a specialist a draft with each exception called out before entry.", "author": "AIBAST", "tags": [ "manufacturing", "order-entry", "purchase-order", "quote-validation", "sales-operations", "erp" ], "category": "manufacturing", "quality_tier": "verified", "requires_env": [], "dependencies": [ "@rapp/basic-agent" ], "_file": "agents/@aibast-agents-library/manufacturing_stacks/order_entry_validation_stack/order_entry_validation_agent.py", "_size_kb": 23.7, "_lines": 424, "_sha256": "8567a365f6975adfd7c995fcfb95425d1f065e9534d0e5c968237ac7b9713a89", "_install_prefix": "order_entry_validation__", "_install_filename": "order_entry_validation__8567a365f697_agent.py", "_stack": "order_entry_validation", "_stack_vertical": "manufacturing", "_synthetic_data": true, "_added_at": "2026-10-08T10:30:56-04:00", "_solution": { "advertised_name": "Order Entry Validation Agent", "sharepoint_list_name": "Order Entry Validation Agent", "slot": 54, "executive_summary": "Turn inbound customer purchase orders into validated draft sales orders: extract every PO field, check prices and terms against the accepted quote, catch invalid product configurations, classify the order, and hand a specialist a draft with each exception called out before entry.", "industries": [ "Manufacturing" ], "personas": [ "Sales Operations Specialists", "Sales Operations Managers", "Product Configuration Analysts" ], "featured_tools": [ "Microsoft Outlook", "Microsoft Teams", "Dynamics 365" ], "agent_requirements": [ "Microsoft Copilot Studio" ], "capabilities": [ "Reads each PO into structured fields and shows the queue", "Validates the PO against the accepted quote with a per-line price tolerance", "Checks the configuration and classifies the order", "Prepares a draft sales order with every exception for the specialist" ], "outcomes": [ "Validates orders without manual re-keying", "Stops errors before they reach the ERP", "Enters ready orders first and routes exceptions with a reason" ], "customer_scenario": [ "Re-keying PO fields is slow and error-prone", "Price and terms mismatches against the quote are missed", "Incompatible or incomplete configurations reach the ERP", "Credits, rework and late shipments follow" ], "slide_title": "Order Entry Validation Agent", "slide_summary": "Turns inbound purchase orders into validated draft sales orders, checking prices, terms and configurations against the quote before entry.", "scenario_name": "A manufacturer of portable instruments", "challenge_intro": "Sales operations enters every order by hand:", "actions_intro": "The Order Entry Validation Agent:", "outcomes_intro": "With the agent the team:", "opportunity_statements": [ "Purchase orders arrive in many formats and are re-keyed by hand", "Prices and terms are checked against the quote line by line", "Invalid configurations slip into the ERP", "Exceptions are found after the order ships" ], "module_name": null, "module_description": null, "source_verified": true, "sales_headline": "Turn inbound purchase orders into validated draft sales orders", "card_pitch": "Give a sales operations team one assistant that reads each customer purchase order, checks prices and terms against the accepted quote, catches invalid product configurations, classifies the order, and hands the specialist a draft sales order with every exception called out.", "why_try": "The demo works a five-order queue in seven plain questions, from reading a PO to a ready-to-enter draft, and nothing is ever submitted to the ERP.", "customer_challenge_copy": "Sales operations specialists re-key every purchase order by hand and compare it to the quote line by line, so price and terms mismatches and invalid configurations slip into the ERP and cause credits, rework and late shipments.", "microsoft_ai_story": "Microsoft Copilot Studio hosts the conversation in Microsoft Teams or Microsoft 365 Copilot; approved connections to the shared PO mailbox in Outlook, the CRM quote in Dynamics 365 and the ERP can be bound later with least-privilege access and human approval.", "business_value": [ "Cuts manual re-keying and line-by-line quote comparison for every purchase order.", "Stops price, terms and configuration errors before they reach the ERP.", "Lets specialists enter ready orders first and route exceptions with a named reason." ], "search_terms": [ "Dynamics 365", "ERP", "Microsoft Copilot Studio", "Microsoft Outlook", "Microsoft Teams", "Order Entry Validation Agent", "Product Configuration Analyst", "Sales Operations Manager", "Sales Operations Specialist", "configuration_check", "draft_sales_order", "extract_po", "order entry", "order_classification", "order_queue", "product configuration", "purchase order", "queue_review", "quote validation", "quote_validation", "sales operations" ], "journey_stage": "Optimize", "blueprint_role": "Establishes a governed validation layer between the inbound PO and the ERP that separates extraction and checks from order submission and customer contact.", "sample_prompts": [ { "label": "Sales Operations Specialist: Pending PO queue", "prompt": "What's in my PO queue this morning?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=order-entry-validation-order-queue" }, { "label": "Sales Operations Specialist: Purchase order extraction", "prompt": "Read PO-4471 and pull out the order details.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=order-entry-validation-extract-po" }, { "label": "Sales Operations Specialist: Quote validation", "prompt": "Does PO-4471 match the quote?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=order-entry-validation-quote-validation" }, { "label": "Product Configuration Analyst: Configuration check", "prompt": "Check the product configuration on the Coho Refining order.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=order-entry-validation-configuration-check" }, { "label": "Sales Operations Specialist: Order classification", "prompt": "What order type, class and sales channel should PO-4471 go in as?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=order-entry-validation-order-classification" }, { "label": "Sales Operations Specialist: Draft sales order", "prompt": "Prepare the sales order for PO-4471 so I can enter it.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=order-entry-validation-draft-sales-order" }, { "label": "Sales Operations Manager: Queue readiness review", "prompt": "Across the whole queue, which orders are ready to enter and which need a fix?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=order-entry-validation-queue-review" } ], "architecture": { "business_flow": [ "Customer purchase order arrives", "Microsoft Teams or Microsoft 365 Copilot", "Order Entry Validation Agent", "Accepted quote and configuration rule book", "Draft sales order with exceptions", "Specialist enters the order in the ERP" ], "capabilities": [ { "operation": "order_queue", "name": "Pending PO queue", "purpose": "Use when someone asks what is in the purchase-order queue." }, { "operation": "extract_po", "name": "Purchase order extraction", "purpose": "Use when someone asks to read a purchase order or pull out its fields." }, { "operation": "quote_validation", "name": "Quote validation", "purpose": "Use when someone asks whether a PO matches its quote, prices or terms." }, { "operation": "configuration_check", "name": "Configuration check", "purpose": "Use when someone asks to check an order's product configuration, kits or pairings." }, { "operation": "order_classification", "name": "Order classification", "purpose": "Use when someone asks which order type, order class or sales channel applies." }, { "operation": "draft_sales_order", "name": "Draft sales order", "purpose": "Use when someone asks to prepare or draft the sales order for a PO." }, { "operation": "queue_review", "name": "Queue readiness review", "purpose": "Use when someone asks which orders across the queue are ready to enter and which need a fix." } ], "easy_mode": [ "Verify Power Platform CLI 2.9.3 or newer and authenticate to the approved target environment.", "Initialize a modern cli-copilot project with an approved publisher prefix and preserve the generated identity settings.", "Upload the two packaged synthetic knowledge files and every operation-specific SKILL.md.", "Bind only approved least-privilege production connections after security, privacy, and business-owner review.", "Push the Draft agent, replay every locked persona case, and stop before publish for explicit approval." ], "local_install_prompt": "Install Order Entry Validation Agent from https://raw.githubusercontent.com/microsoft/aibast-agents-library/main/solutions/order-entry-validation/deployment.json, verify the expected tool and smoke case, and preserve the synthetic read-only boundary. Do not ask me to open a terminal. Do not connect customer systems or execute external actions.", "copilot_studio_prompt": "Using the Microsoft Copilot Studio plugin, initialize a Draft agent for Order Entry Validation Agent from https://raw.githubusercontent.com/microsoft/aibast-agents-library/main/solutions/order-entry-validation/deployment.json. Upload the two synthetic knowledge files and 7 skills, replay every locked case, and Stop before publish. Do not bind production connections without explicit customer approval.", "required_connections": [ "Outlook shared PO mailbox (read only)", "Dynamics 365 quotes (read only)", "ERP sales order entry (specialist only)" ], "manual_commands": [ "pac auth create", "pac copilot init --name \"Order Entry Validation Agent\" --publisher-prefix --authoring-mode cli-copilot --project-dir \"\" --environment \"\"", "pac connection list --environment \"\"", "pac copilot pull --project-dir \"\"", "pac copilot push --project-dir \"\"", "pac copilot publish --bot \"\" --environment \"\"" ], "acceptance_checks": [ "Every exposed operation has one persona-language case and one SKILL.md.", "Exactly two synthetic knowledge files are present.", "Responses state the relevant human approval or review boundary.", "No operation claims an external write, communication, commitment, or protected decision.", "Publishing remains a separate user-approved step." ], "hard_mode": [ "Run pac copilot init and verify settings.mcs.yml before authoring components.", "Author global instructions that preserve the packaged synthetic-only and human-approval boundaries.", "Add the two knowledge files and one skill for every exposed operation.", "Create dedicated connection references only for approved systems and verify least-privilege access.", "Pull before push, inspect every generated file, replay every locked case, and stop before publish." ] }, "curated_copy": true, "source": "field", "origin": "both", "onepager": "54. Order Entry Validation Agent one-pager.pptx", "demo_video": "54-Order Entry Validation Agent.mp4", "has_onepager": true, "has_demo_video": true, "promise_coverage": 100, "is_primary": true, "aliases": [], "package": { "slug": "order-entry-validation", "quest_url": "solutions/order-entry-validation/quest.html", "manual_tutorial_url": "solutions/order-entry-validation/manual-tutorial.html", "export_manifest_url": "solutions/order-entry-validation/export-manifest.json" } }, "_demo": { "slug": "order-entry-validation", "case_count": 7, "personas": [ "Product Configuration Analyst", "Sales Operations Manager", "Sales Operations Specialist" ], "prompts": [ "What's in my PO queue this morning?", "Read PO-4471 and pull out the order details.", "Does PO-4471 match the quote?" ] }, "_catalog_kind": "solution", "_readiness": [ "Demo proven", "Conversation tested", "Synthetic data", "Production pattern" ] }, { "schema": "rapp-agent/1.0", "name": "@aibast-agents-library/order-status-communication", "version": "1.1.0", "display_name": "Order Status Communications Agent", "description": "Analyze a fixed synthetic order snapshot and prepare review-ready status, delay, recovery, and customer-message drafts. Never send a customer update or change an order, shipment, or production schedule.", "author": "AIBAST", "tags": [ "orders", "communication", "shipment", "customer-service", "manufacturing" ], "category": "manufacturing", "quality_tier": "verified", "requires_env": [], "dependencies": [ "@rapp/basic-agent" ], "_file": "agents/@aibast-agents-library/manufacturing_stacks/order_status_communication_stack/order_status_communication_agent.py", "_size_kb": 31.0, "_lines": 637, "_sha256": "1c7d61f65444699f15d5689ace5e6ffe18ca390db030c551b3aec760a2d4f86c", "_install_prefix": "order_status_communication__", "_install_filename": "order_status_communication__1c7d61f65444_agent.py", "_stack": "order_status_communication", "_stack_vertical": "manufacturing", "_synthetic_data": true, "_added_at": "2026-03-18T13:23:12-04:00", "_solution": { "advertised_name": "Order Status Communications Agent", "sharepoint_list_name": "Order Status & Delay Agent", "slot": 42, "executive_summary": "Automate proactive order updates, delay management, and customer communications to protect relationships and revenue.", "industries": [ "Manufacturing" ], "personas": [ "Customer Service Rep", "Account Manager", "Operations Leader" ], "featured_tools": [ "Dynamics 365", "Microsoft Teams" ], "agent_requirements": [ "Microsoft Copilot Studio" ], "capabilities": [ "Identify delays proactively with automated monitoring", "Generate recovery plans like schedule adjustments, alternative sourcing, and expedited logistics", "Communicate updates across channels with consistent messaging", "Tailor notifications and actions to the customer tier and preferences" ], "outcomes": [ "Reduced delivery delays with faster recovery plan activation", "Gave timely, transparent updates that increased customer trust", "Minimized inquiry volume with proactive updates and next steps", "Preserved account health and supplier ratings through consistent disruption handling" ], "customer_scenario": [ "Customers learned about delays too late, triggering escalations", "Manual status reporting consumed significant customer service time", "Cross\u2011functional coordination was slow and inconsistent", "Maintaining customer trust during disruptions was increasingly difficult" ], "slide_title": "Order Status Communications Agent", "slide_summary": "Automates proactive order updates, delay management, and customer communications to protect relationships and revenue.", "scenario_name": "An automotive supplier", "challenge_intro": "An automotive supplier faced issues with monitoring and communicating production and supply delays:", "actions_intro": "The supplier used the Order Status Communications Agent to:", "outcomes_intro": "With the agent, the supplier:", "opportunity_statements": [ "Automates proactive order updates, delay management, and customer communications to protect relationships and revenue.", "Monitor order and production status in real time to detect delays early and prevent customer surprises", "Automate tailored, multi\u2011channel communications that keep customers informed and reduce inbound inquiries", "Generate and execute recovery plans instantly to minimize disruption and preserve customer confidence" ], "module_name": null, "module_description": null, "source_verified": true, "sales_headline": "Align on the order story before the customer receives it", "card_pitch": "Bring order, production, shipment, delay, recovery, owner, and channel context into one governed review and prepare consistent customer-message drafts without changing operations or sending updates.", "why_try": "It makes service communication AI concrete: customer service and account teams can identify at-risk orders, validate shipment evidence, review recovery context, and prepare tailored drafts with an authorized sender still in control.", "customer_challenge_copy": "A manufacturer assembles order status, production progress, shipment evidence, delay causes, recovery options, customer preferences, and account ownership from disconnected systems. Manual coordination slows communication and can create inconsistent or unapproved commitments.", "microsoft_ai_story": "Microsoft Copilot Studio provides the governed conversational experience. Dynamics 365 Supply Chain Management and approved MES sources contribute order and production context; carrier integrations contribute shipment evidence; Dynamics 365 Customer Service contributes account and preference context; Power Automate and Microsoft Teams support reviewed communications and approvals.", "business_value": [ "Improves shared visibility into order, shipment, delay, recovery, ownership, and customer-preference context.", "Accelerates preparation of consistent customer updates while keeping commitments grounded in approved evidence.", "Protects customer trust by separating draft generation from order changes, schedule changes, logistics action, and outbound communication." ], "search_terms": [ "order status communication", "customer service representative", "account manager", "manufacturing order status", "shipment tracking", "delay notification", "customer update draft", "recovery plan", "Dynamics 365 Customer Service", "Dynamics 365 Supply Chain Management", "Power Automate communication", "Copilot Studio manufacturing" ], "journey_stage": "Serve", "blueprint_role": "Creates a governed communication layer that connects order, production, logistics, recovery, account, and channel evidence while preserving operational systems and authorized-sender controls.", "sample_prompts": [ { "label": "Review orders needing attention", "prompt": "Which orders are on track or delayed, and where should customer service focus its review?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=order-status-review" }, { "label": "Validate recorded shipment evidence", "prompt": "What shipment evidence is recorded for the shipped order, and what still needs carrier validation?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=order-shipment-tracking" }, { "label": "Prepare an internal delay review", "prompt": "Prepare the internal delay and recovery review for the at-risk customer order without changing any schedule.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=order-delay-review" }, { "label": "Draft customer updates for approval", "prompt": "Draft the customer updates for approval, but do not send an email, portal update, EDI message, or Teams message.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=order-customer-update" } ], "architecture": { "business_flow": [ "Customer service representatives, account managers, and operations leaders", "Microsoft 365 Copilot or Copilot Studio", "Order Status Communications Agent", "Dynamics 365 Supply Chain Management and approved MES data", "Carrier and logistics tracking", "Dynamics 365 Customer Service", "Authorized communication reviewers and senders" ], "capabilities": [ { "operation": "order_lookup", "name": "Order status and risk overview", "purpose": "Summarizes fixed order progress, promise, value, and delay evidence for internal review." }, { "operation": "shipment_tracking", "name": "Shipment evidence review", "purpose": "Reports the recorded carrier and route evidence while requiring validation in the approved tracking system." }, { "operation": "delay_notification", "name": "Internal delay and recovery review", "purpose": "Brings delay reason, recovery options, account ownership, response expectations, and preferred channel together without changing an order or schedule." }, { "operation": "customer_update", "name": "Customer communication drafting", "purpose": "Prepares customer-specific message drafts for approval without sending email, EDI, portal, Teams, or other updates." } ], "easy_mode": [ "Verify Power Platform CLI 2.9.3 or newer and authenticate to the target environment.", "Use the Copilot Studio plugin initialization workflow to create a modern, sync-connected agent project with an approved publisher prefix.", "Give the Copilot Studio Architect this blueprint and the initialized project so it writes governed instructions, operation-specific skills, and approval-aware tool definitions.", "Bind approved order, production, carrier, logistics, CRM, communication, and collaboration connections through dedicated connection references.", "Use the Copilot Studio Manage workflow to push the project, replay every locked prompt, confirm that no operational change or customer update is claimed, and stop before publishing." ], "local_install_prompt": "You are GitHub Copilot running in Agent mode in VS Code. Install and open the AI BAST Order Status Communications Agent from this deployment recipe: https://raw.githubusercontent.com/microsoft/aibast-agents-library/main/solutions/order-status-communication/deployment.json. Own the entire setup. Do not ask me to open a terminal, run a command, clone a repository, or install the Brainstem myself. Inspect the operating system; check the recipe health endpoint; if the Brainstem is missing, run the official installer yourself using the recipe's OS-appropriate installer; install or refresh the policy-clean launcher; fetch registry.json and the raw agent source; verify the file against the registry SHA-256; place it under the discovered Brainstem agents directory with the recipe target filename; start the Brainstem; verify /health reports the expected tool; run the recipe smoke prompt through /chat and confirm the expected agent fired; then open the local chat UI in my browser. If authentication requires a browser confirmation, launch the supported device flow and ask me only to complete that browser step. Do not publish, connect customer systems, change orders, production schedules, shipments, recovery plans, or send customer communications, or silently substitute another agent. If any step fails, stop and show the exact failed step and output.", "copilot_studio_prompt": "You are GitHub Copilot running in Agent mode in VS Code with the Microsoft Copilot Studio plugin installed. Promote the AI BAST Order Status Communications Agent using this deployment recipe: https://raw.githubusercontent.com/microsoft/aibast-agents-library/main/solutions/order-status-communication/deployment.json. Own every command and file change; do not ask me to use a terminal. Ask me only for the target Power Platform environment ID and publisher prefix. Verify PAC CLI is at least 2.9.3, authenticate if needed, initialize a new cli-copilot project, and use the Copilot Studio Architect workflow to author modern YAML that preserves order_lookup, shipment_tracking, delay_notification, and customer_update. Bind approved Dynamics 365 Supply Chain Management, MES, carrier, logistics, Dynamics 365 Customer Service, Power Automate, email, EDI, portal, and Microsoft Teams connections. Review every generated YAML file and connection, pull before push, run pac copilot push, and replay every locked prompt. Confirm that order, schedule, shipment, sourcing, logistics, recovery, and outbound communication actions remain unavailable without explicit authorization and action tools. Stop before publish and show the generated files, validation evidence, unresolved connection work, and the exact publish command that would make it live.", "required_connections": [ "Dynamics 365 Supply Chain Management or another approved order source", "Approved MES, carrier, and logistics tracking", "Dynamics 365 Customer Service or another approved CRM", "Power Automate and approved email, EDI, portal, or Teams communication channels" ], "manual_commands": [ "pac auth create", "pac copilot init --name \"Order Status Communications Agent\" --publisher-prefix --authoring-mode cli-copilot --project-dir \"\" --environment \"\"", "pac connection list --environment \"\"", "pac copilot pull --project-dir \"\"", "pac copilot push --project-dir \"\"", "pac copilot publish --bot \"\" --environment \"\"" ], "acceptance_checks": [ "settings.mcs.yml exists and preserves the initialized identity, environment binding, language, and authoring mode.", "Every implemented order-communication operation is represented by instructions, an uploadable skill, or a bound tool.", "Both synthetic knowledge files are loaded and clearly label every exact figure and identifier as fictional pilot evidence.", "Order, production, carrier, logistics, CRM, communication, and collaboration sources use dedicated and verified connection references.", "The order, shipment, delay-review, and customer-draft prompts return the expected synthetic records.", "No output claims an order, schedule, shipment, recovery plan, sourcing action, logistics action, or customer communication was changed or sent.", "Public value statements remain qualitative and publishing remains a separate, user-approved step." ], "hard_mode": [ "Run pac copilot init and confirm settings.mcs.yml exists.", "Author global instructions that separate order evidence, recovery context, draft generation, communication approval, and external sending.", "Upload the two synthetic Markdown knowledge files and every operation-specific SKILL.md file from this package.", "Create read tools for approved order, production, carrier, logistics, CRM, and preference data with dedicated connection references.", "Implement explicit approval and action tools for any future order, schedule, shipment, recovery, sourcing, logistics, or communication change; keep them unavailable in the draft-only pilot.", "Validate every component schema and filename, run pac copilot push, inspect the generated agent, replay all locked prompts, and stop before publish." ] }, "curated_copy": true, "source": "catalog", "origin": null, "onepager": "44. Order Status Communications Agent one-pager.pptx", "demo_video": "44-Order Status Communication Agent.mp4", "has_onepager": true, "has_demo_video": true, "promise_coverage": 70, "is_primary": true, "aliases": [], "package": { "slug": "order-status-communication", "quest_url": "solutions/order-status-communication/quest.html", "manual_tutorial_url": "solutions/order-status-communication/manual-tutorial.html", "export_manifest_url": "solutions/order-status-communication/export-manifest.json" } }, "_demo": { "slug": "order-status-communication", "case_count": 7, "personas": [ "Account Manager", "Customer Service Representative", "Operations Leader", "Production Manager" ], "prompts": [ "Which orders are on track or delayed, and where should customer service focus its review?", "What shipment evidence is recorded for the shipped order, and what still needs carrier validation?", "Prepare the internal delay and recovery review for the E-Cars Corp transmission housing order without changing any schedule." ] }, "_catalog_kind": "solution", "_readiness": [ "Demo proven", "Conversation tested", "Synthetic data", "Production pattern" ] }, { "schema": "rapp-agent/1.0", "name": "@aibast-agents-library/production-line-optimization", "version": "1.2.0", "display_name": "Product Line Optimization Agent", "description": "Provide intelligent production capacity analysis and optimization planning to boost throughput and efficiency while maintaining quality.", "author": "AIBAST", "tags": [ "production", "OEE", "bottleneck", "throughput", "manufacturing" ], "category": "manufacturing", "quality_tier": "verified", "requires_env": [], "dependencies": [ "@rapp/basic-agent" ], "_file": "agents/@aibast-agents-library/manufacturing_stacks/production_line_optimization_stack/production_line_optimization_agent.py", "_size_kb": 31.1, "_lines": 651, "_sha256": "fa16e4751cdddbdbbe8a683af320a8b22a5bad166d42765d973ffb0a150524c2", "_install_prefix": "production_line_optimization__", "_install_filename": "production_line_optimization__fa16e4751cdd_agent.py", "_stack": "production_line_optimization", "_stack_vertical": "manufacturing", "_synthetic_data": true, "_added_at": "2026-03-18T13:23:12-04:00", "_solution": { "advertised_name": "Product Line Optimization Agent", "sharepoint_list_name": "Product Line Optimization Agent", "slot": 1, "executive_summary": "Provide intelligent production capacity analysis and optimization planning to boost throughput and efficiency while maintaining quality.", "industries": [ "Manufacturing" ], "personas": [ "Plant Manager", "Production Engineer", "Operations Director" ], "featured_tools": [ "Dynamics 365 for ERP", "Power BI", "Azure IoT Hub" ], "agent_requirements": [ "Microsoft Copilot Studio" ], "capabilities": [ "Perform real-time analysis on key production metrics across stations and lines", "Identify and propose solutions to throughput constraints", "Plan to optimize processes, crew allocation, and investments", "Scale capacity to meet the seasonal demand surge" ], "outcomes": [ "Increased line capacity and reduced cycle times through targeted optimizations", "Higher Overall Equipment Effectiveness (OEE) and throughput, exceeding holiday targets", "Consistent high quality and increased ROI on investments" ], "customer_scenario": [ "Production lines were limited by a bottleneck at a key station, restricting daily output", "Manual scheduling and lack of real-time analytics limited timely response to demand spikes", "Capacity limits were reached, putting quality and timely delivery at risk" ], "slide_title": "Production Line Optimization Agent", "slide_summary": "Provide intelligent production capacity analysis and optimization planning to boost throughput and efficiency while maintaining quality.", "scenario_name": "A consumer electronics manufacturer", "challenge_intro": "The manufacturer struggled to keep up with a holiday demand surge.", "actions_intro": "Used the Product Line Optimization Agent to:", "outcomes_intro": "With the Agent, the manufacturer was able to achieve:", "opportunity_statements": [ "Analyze production line performance in real time to identify bottlenecks and operational inefficiencies", "Recommend targeted equipment and process improvements to improve output quality and cycle times", "Plan and coordinates implementation, minimizing disruption and providing a high return on investment" ], "module_name": null, "module_description": null, "source_verified": true, "sales_headline": "Find the constraint before it becomes the production plan", "card_pitch": "Turn line efficiency, station cycle time, quality, and shift data into an operating view that exposes bottlenecks, compares improvement paths, and gives plant leaders a practical plan for protecting throughput and quality.", "why_try": "It makes operational AI concrete: a plant manager can ask where output is being lost, see the constraining station, and evaluate process, equipment, quality, or staffing responses.", "customer_challenge_copy": "A manufacturer facing demand pressure lacks one shared view of line performance. Bottlenecks, shift constraints, and defect patterns are reviewed manually, making it harder to respond without trading away quality.", "microsoft_ai_story": "Microsoft Copilot Studio provides the natural-language experience. Dynamics 365 ERP contributes production and resource context, Azure IoT Hub supplies equipment signals, and Power BI presents line health and improvement scenarios.", "business_value": [ "Improves visibility into where capacity and quality are being lost across production lines.", "Accelerates bottleneck analysis and turns operational signals into targeted improvement options.", "Strengthens shift planning and investment decisions while keeping quality and disruption risk visible." ], "search_terms": [ "production line optimization", "plant manager", "OEE", "bottleneck analysis", "cycle time", "takt time", "throughput planning", "quality defects", "shift planning", "Dynamics 365 ERP", "Azure IoT Hub", "Power BI manufacturing", "Copilot Studio" ], "journey_stage": "Optimize", "blueprint_role": "Creates the operational intelligence layer for a manufacturing AI roadmap by connecting equipment signals, ERP context, quality evidence, and improvement planning.", "sample_prompts": [ { "label": "See which production line needs attention", "prompt": "Which production line needs attention today, and what is driving the loss?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=production-line-health" }, { "label": "Find the constraining station", "prompt": "Where is the bottleneck on each line, and which station should the plant team address first?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=production-bottleneck" }, { "label": "Compare throughput improvement options", "prompt": "Give me the practical options to improve throughput without hiding the quality tradeoffs.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=production-options" }, { "label": "Build the three-shift production plan", "prompt": "Translate the current line performance into a day, swing, and night shift plan.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=production-shift-plan" } ], "architecture": { "business_flow": [ "Plant leaders and production engineers", "Microsoft 365 Copilot or Copilot Studio", "Product Line Optimization Agent", "Dynamics 365 ERP", "Azure IoT Hub", "Power BI", "Production and maintenance teams" ], "capabilities": [ { "operation": "line_efficiency", "name": "Line health overview", "purpose": "Combines availability, performance, quality, and output context to show where operating attention is needed." }, { "operation": "bottleneck_analysis", "name": "Station-level constraint analysis", "purpose": "Compares cycle time with takt time and quality signals to identify the station constraining each line." }, { "operation": "throughput_optimization", "name": "Improvement option planning", "purpose": "Frames process, equipment, and quality interventions so plant leaders can compare practical response paths." }, { "operation": "shift_planning", "name": "Shift production planning", "purpose": "Translates line performance and staffing context into a shift-level operating plan." } ], "easy_mode": [ "Verify Power Platform CLI 2.9.3 or newer and authenticate to the target environment.", "Use the Copilot Studio plugin initialization workflow to create a modern, sync-connected agent project with an approved publisher prefix.", "Give the Copilot Studio Architect this blueprint and the initialized project so it writes the instructions, analysis skills, and tool definitions as YAML.", "Bind Dynamics 365 ERP and approved operational data actions; expose Azure IoT Hub signals through a supported connector, workflow, or API layer; link Power BI for the visual reporting surface.", "Use the Copilot Studio Manage workflow to run pac copilot push, then validate the three canned prompts before publishing." ], "local_install_prompt": "You are GitHub Copilot running in Agent mode in VS Code. Install and open the AI BAST Product Line Optimization Agent from this deployment recipe: https://raw.githubusercontent.com/microsoft/aibast-agents-library/main/solutions/product-line-optimization/deployment.json. Own the entire setup. Do not ask me to open a terminal, run a command, clone a repository, or install the Brainstem myself. Inspect the operating system; check the recipe health endpoint; if the Brainstem is missing, run the official installer yourself using the recipe's OS-appropriate installer; install or refresh the policy-clean launcher; fetch registry.json and the raw agent source; verify the file against the registry SHA-256; place it under the discovered Brainstem agents directory with the recipe target filename; start the Brainstem; verify /health reports the expected tool; run the recipe smoke prompt through /chat and confirm the expected agent fired; then open the local chat UI in my browser. If authentication requires a browser confirmation, launch the supported device flow and ask me only to complete that browser step. Do not publish, connect customer systems, or silently substitute another agent. If any step fails, stop and show the exact failed step and output.", "copilot_studio_prompt": "You are GitHub Copilot running in Agent mode in VS Code with the Microsoft Copilot Studio plugin installed. Promote the AI BAST Product Line Optimization Agent using this deployment recipe: https://raw.githubusercontent.com/microsoft/aibast-agents-library/main/solutions/product-line-optimization/deployment.json. Own every command and file change; do not ask me to use a terminal. Ask me only for the target Power Platform environment ID and publisher prefix. Verify PAC CLI is at least 2.9.3, authenticate if needed, initialize a new cli-copilot project, and use the Copilot Studio Architect workflow to author modern YAML that preserves line_efficiency, bottleneck_analysis, throughput_optimization, and shift_planning. Bind Dynamics 365 ERP, expose Azure IoT Hub telemetry through an approved connector, workflow, or API, and link Power BI as the reporting surface. Review every generated YAML file and connection, pull before push, run pac copilot push, and replay the three recipe acceptance prompts. Stop before publish and show me the generated files, validation evidence, unresolved connection work, and the exact publish command that would make it live.", "required_connections": [ "Dynamics 365 ERP", "Azure IoT Hub through an approved connector, workflow, or API", "Power BI" ], "manual_commands": [ "pac auth create", "pac copilot init --name \"Product Line Optimization Agent\" --publisher-prefix --authoring-mode cli-copilot --project-dir \"\" --environment \"\"", "pac connection list --environment \"\"", "pac copilot pull --project-dir \"\"", "pac copilot push --project-dir \"\"", "pac copilot publish --bot \"\" --environment \"\"" ], "acceptance_checks": [ "settings.mcs.yml exists and preserves the initialized identity, environment binding, language, and authoring mode.", "Every implemented production operation is represented by instructions, a skill, or a bound tool.", "ERP, telemetry, and reporting connections use dedicated and verified connection references.", "The line-health, bottleneck, improvement-option, and shift-planning prompts produce the expected synthetic records and decisions.", "Synthetic estimates are labeled as planning evidence rather than customer performance commitments.", "Publishing remains a separate, user-approved step after push and validation." ], "hard_mode": [ "Run pac copilot init and confirm settings.mcs.yml exists.", "Author instructions that define line-health reasoning, bottleneck selection, option comparison, quality guardrails, and shift-planning behavior.", "Create tools for ERP production context, IoT telemetry retrieval, and any approved workflow or API used to expose equipment signals.", "Add Power BI as the reporting destination or linked analytical surface, and verify every connection reference in the target environment.", "Validate component schemas and filenames, run pac copilot push, inspect the generated agent in Copilot Studio, and replay the canned prompts." ] }, "curated_copy": true, "source": "catalog", "origin": null, "onepager": "Product Line Optimization Agent one-pager.pptx", "demo_video": "#1-Product Line Optimization Agent.mp4", "has_onepager": true, "has_demo_video": true, "promise_coverage": 85, "is_primary": true, "aliases": [], "package": { "slug": "product-line-optimization", "quest_url": "solutions/product-line-optimization/quest.html", "manual_tutorial_url": "solutions/product-line-optimization/manual-tutorial.html", "export_manifest_url": "solutions/product-line-optimization/export-manifest.json" } }, "_demo": { "slug": "product-line-optimization", "case_count": 4, "personas": [ "Operations Director", "Plant Manager", "Production Engineer" ], "prompts": [ "Which production line needs attention today, and what is driving the loss?", "Where is the bottleneck on each line, and which station should the plant team address first?", "Give me the practical options to improve throughput without hiding the quality tradeoffs." ] }, "_catalog_kind": "solution", "_readiness": [ "Demo proven", "Conversation tested", "Synthetic data", "Production pattern" ] }, { "schema": "rapp-agent/1.0", "name": "@aibast-agents-library/supplier-risk-monitoring", "version": "1.0.0", "display_name": "Supply Risk Monitoring Agent", "description": "Analyze a fixed synthetic supplier-risk snapshot and recommend review-ready mitigation options. Never contact suppliers, change sourcing, place orders, or approve procurement action.", "author": "AIBAST", "tags": [ "supplier", "risk", "procurement", "supply-chain", "manufacturing" ], "category": "manufacturing", "quality_tier": "verified", "requires_env": [], "dependencies": [ "@rapp/basic-agent" ], "_file": "agents/@aibast-agents-library/manufacturing_stacks/supplier_risk_monitoring_stack/supplier_risk_monitoring_agent.py", "_size_kb": 29.9, "_lines": 651, "_sha256": "5711192300f2f6311f668883879be4b9a424fb982dc37ac49b18205536dd9322", "_install_prefix": "supplier_risk_monitoring__", "_install_filename": "supplier_risk_monitoring__5711192300f2_agent.py", "_stack": "supplier_risk_monitoring", "_stack_vertical": "manufacturing", "_synthetic_data": true, "_added_at": "2026-03-18T13:23:12-04:00", "_solution": { "advertised_name": "Supply Risk Monitoring Agent", "sharepoint_list_name": "Supply Risk Intelligence Agent", "slot": 20, "executive_summary": "Deliver real-time risk intelligence and planning to protect production continuity and reduce disruption exposure.", "industries": [ "Manufacturing" ], "personas": [ "Procurement Manager", "Supply Chain Director" ], "featured_tools": [ "Dynamics 365", "Microsoft Teams" ], "agent_requirements": [ "Microsoft Copilot Studio" ], "capabilities": [ "Unify risk data and score suppliers in real time", "Provide visibility into financial health, operational capacity, and logistics reliability", "Recommend mitigation actions like safety stock and dual sourcing", "Model cost exposure and ROI" ], "outcomes": [ "Reduced disruption exposure with early risk detection and mitigation", "Protected revenue by preventing supply chain interruptions", "Improved procurement agility with real-time insights", "Strengthened supplier relationships with better risk communication and proactive planning" ], "customer_scenario": [ "Components sourced from high-risk regions with unstable logistics", "Risk signals scattered across ERP, supplier portals, and financial feeds", "Manual analysis delayed response and increased disruption exposure", "Rising costs and margin erosion from reactive mitigation efforts" ], "slide_title": "Supply Risk Monitoring Agent", "slide_summary": "Deliver real-time risk intelligence and planning to protect production continuity and reduce disruption exposure.", "scenario_name": "A discrete manufacturing company", "challenge_intro": "A manufacturer faced supplier risk that threatened production continuity:", "actions_intro": "The manufacturer used the Supply Risk Monitoring Agent to:", "outcomes_intro": "With the Agent, the firm:", "opportunity_statements": [ "Perform real-time risk analysis on financial, operational, and logistics indicators for proactive risk detection", "Recommend proactive mitigation strategies that reduce disruption impact and safeguard revenue", "Provide continuous monitoring and AI-driven alerts for high visibility and rapid response to emerging risks" ], "module_name": null, "module_description": null, "source_verified": true, "sales_headline": "Turn scattered supplier signals into an explainable continuity review", "card_pitch": "Unify supplier quality, delivery, financial, geopolitical, incident, and backup-option evidence so procurement leaders can prioritize review without contacting, qualifying, selecting, or ordering from suppliers.", "why_try": "It makes supplier-risk AI concrete: leaders can ask where exposure is concentrated, understand why a relationship is elevated, and compare mitigation options with procurement authority preserved.", "customer_challenge_copy": "A manufacturer reviews supplier evidence across ERP, supplier-management, logistics, quality, and external-risk systems. Fragmented analysis delays continuity planning and can confuse recommendations with authorized sourcing action.", "microsoft_ai_story": "Microsoft Copilot Studio provides the governed conversational experience. Dynamics 365 Supply Chain Management contributes supplier, order, quality, and inventory context; approved financial and logistics feeds contribute risk signals; Power BI presents portfolio exposure; and Microsoft Teams supports procurement review and escalation.", "business_value": [ "Improves visibility into supplier relationships and incidents that deserve continuity review.", "Accelerates explanation of risk by connecting commercial, operational, financial, logistics, and geopolitical evidence.", "Strengthens procurement governance by separating mitigation recommendations from supplier outreach, qualification, allocation, and ordering." ], "search_terms": [ "supplier risk monitoring", "supply risk", "procurement manager", "supply chain director", "supplier scorecard", "supplier disruption", "financial risk", "logistics risk", "dual sourcing", "alternative sourcing", "Dynamics 365 Supply Chain Management", "Power BI supplier risk", "Copilot Studio manufacturing" ], "journey_stage": "Protect", "blueprint_role": "Creates a governed supplier-risk intelligence layer that connects operational and external evidence to mitigation review while preserving procurement, qualification, contracting, and ordering authority.", "sample_prompts": [ { "label": "Review concentrated supplier exposure", "prompt": "Where is supplier exposure concentrated, and which relationships should leadership review first?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=supplier-risk-dashboard" }, { "label": "Explain an elevated supplier", "prompt": "Explain why TechnoCore is elevated and show me the evidence by risk dimension.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=supplier-risk-scorecard" }, { "label": "Review continuity threats", "prompt": "Which recorded disruptions could threaten continuity, and what exposure should we validate?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=supplier-disruption-alerts" }, { "label": "Compare backup sourcing options", "prompt": "Compare backup sourcing options for review, but do not contact, qualify, select, or order from any supplier.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=supplier-alternative-sourcing" } ], "architecture": { "business_flow": [ "Procurement managers, supply chain directors, and continuity leaders", "Microsoft 365 Copilot or Copilot Studio", "Supply Risk Monitoring Agent", "Dynamics 365 Supply Chain Management", "Approved financial, quality, logistics, and external-risk sources", "Power BI", "Authorized procurement and supplier-management reviewers" ], "capabilities": [ { "operation": "risk_dashboard", "name": "Supplier risk portfolio overview", "purpose": "Ranks the fixed supplier snapshot and shows where commercial exposure and incidents deserve review." }, { "operation": "supplier_scorecard", "name": "Supplier evidence scorecards", "purpose": "Explains quality, delivery, financial, geopolitical, and incident evidence by supplier." }, { "operation": "disruption_alerts", "name": "Continuity disruption review", "purpose": "Summarizes recorded incidents and potential exposure while requiring validation against approved sources." }, { "operation": "alternative_sourcing", "name": "Alternative sourcing option review", "purpose": "Compares synthetic backup options without contacting, qualifying, selecting, allocating, or ordering from a supplier." } ], "easy_mode": [ "Verify Power Platform CLI 2.9.3 or newer and authenticate to the target environment.", "Use the Copilot Studio plugin initialization workflow to create a modern, sync-connected agent project with an approved publisher prefix.", "Give the Copilot Studio Architect this blueprint and the initialized project so it writes governed instructions, operation-specific skills, and approval-aware tool definitions.", "Bind approved ERP, supplier-management, quality, financial, logistics, external-risk, reporting, and collaboration connections through dedicated connection references.", "Use the Copilot Studio Manage workflow to push the project, replay every locked prompt, confirm that no supplier action is claimed, and stop before publishing." ], "local_install_prompt": "You are GitHub Copilot running in Agent mode in VS Code. Install and open the AI BAST Supply Risk Monitoring Agent from this deployment recipe: https://raw.githubusercontent.com/microsoft/aibast-agents-library/main/solutions/supplier-risk-monitoring/deployment.json. Own the entire setup. Do not ask me to open a terminal, run a command, clone a repository, or install the Brainstem myself. Inspect the operating system; check the recipe health endpoint; if the Brainstem is missing, run the official installer yourself using the recipe's OS-appropriate installer; install or refresh the policy-clean launcher; fetch registry.json and the raw agent source; verify the file against the registry SHA-256; place it under the discovered Brainstem agents directory with the recipe target filename; start the Brainstem; verify /health reports the expected tool; run the recipe smoke prompt through /chat and confirm the expected agent fired; then open the local chat UI in my browser. If authentication requires a browser confirmation, launch the supported device flow and ask me only to complete that browser step. Do not publish, connect customer systems, contact, qualify, disqualify, select, allocate, contract with, or order from a supplier, or silently substitute another agent. If any step fails, stop and show the exact failed step and output.", "copilot_studio_prompt": "You are GitHub Copilot running in Agent mode in VS Code with the Microsoft Copilot Studio plugin installed. Promote the AI BAST Supply Risk Monitoring Agent using this deployment recipe: https://raw.githubusercontent.com/microsoft/aibast-agents-library/main/solutions/supplier-risk-monitoring/deployment.json. Own every command and file change; do not ask me to use a terminal. Ask me only for the target Power Platform environment ID and publisher prefix. Verify PAC CLI is at least 2.9.3, authenticate if needed, initialize a new cli-copilot project, and use the Copilot Studio Architect workflow to author modern YAML that preserves risk_dashboard, supplier_scorecard, disruption_alerts, and alternative_sourcing. Bind approved Dynamics 365 Supply Chain Management, supplier-management, quality, financial, logistics, external-risk, Power BI, and Microsoft Teams connections. Review every generated YAML file and connection, pull before push, run pac copilot push, and replay every locked prompt. Confirm that supplier outreach, qualification, selection, allocation, contracting, and ordering remain unavailable without explicit authorized tools. Stop before publish and show the generated files, validation evidence, unresolved connection work, and the exact publish command that would make it live.", "required_connections": [ "Dynamics 365 Supply Chain Management or another approved ERP and supplier-management source", "Approved quality, financial, logistics, and external-risk feeds", "Power BI", "Microsoft Teams" ], "manual_commands": [ "pac auth create", "pac copilot init --name \"Supply Risk Monitoring Agent\" --publisher-prefix --authoring-mode cli-copilot --project-dir \"\" --environment \"\"", "pac connection list --environment \"\"", "pac copilot pull --project-dir \"\"", "pac copilot push --project-dir \"\"", "pac copilot publish --bot \"\" --environment \"\"" ], "acceptance_checks": [ "settings.mcs.yml exists and preserves the initialized identity, environment binding, language, and authoring mode.", "Every implemented supplier-risk operation is represented by instructions, an uploadable skill, or a bound tool.", "Both synthetic knowledge files are loaded and clearly label every exact figure and identifier as fictional pilot evidence.", "ERP, supplier, quality, financial, logistics, risk, reporting, and collaboration sources use dedicated and verified connection references.", "The dashboard, scorecard, disruption, and alternative-sourcing prompts return the expected synthetic records.", "No output claims supplier contact, qualification, selection, allocation, contracting, or ordering.", "Public value statements remain qualitative and publishing remains a separate, user-approved step." ], "hard_mode": [ "Run pac copilot init and confirm settings.mcs.yml exists.", "Author global instructions that separate supplier evidence, risk interpretation, mitigation recommendation, procurement approval, and external action.", "Upload the two synthetic Markdown knowledge files and every operation-specific SKILL.md file from this package.", "Create read tools for approved ERP, supplier, quality, financial, logistics, external-risk, and reporting data with dedicated connection references.", "Implement explicit procurement approval and action tools for any future outreach, qualification, selection, allocation, contracting, or ordering; keep them unavailable in the recommendation-only pilot.", "Validate every component schema and filename, run pac copilot push, inspect the generated agent, replay all locked prompts, and stop before publish." ] }, "curated_copy": true, "source": "catalog", "origin": null, "onepager": "Supply Risk Monitoring Agent one-pager.pptx", "demo_video": "9-Supply Risk Monitoring Agent.mp4", "has_onepager": true, "has_demo_video": true, "promise_coverage": 80, "is_primary": true, "aliases": [], "package": { "slug": "supplier-risk-monitoring", "quest_url": "solutions/supplier-risk-monitoring/quest.html", "manual_tutorial_url": "solutions/supplier-risk-monitoring/manual-tutorial.html", "export_manifest_url": "solutions/supplier-risk-monitoring/export-manifest.json" } }, "_demo": { "slug": "supplier-risk-monitoring", "case_count": 8, "personas": [ "Procurement Manager", "Supply Chain Director" ], "prompts": [ "Where is supplier exposure concentrated, and which relationships should leadership review first?", "Explain why TechnoCore is elevated and show me the evidence by risk dimension.", "Which recorded disruptions could threaten continuity, and what exposure should we validate?" ] }, "_catalog_kind": "solution", "_readiness": [ "Demo proven", "Conversation tested", "Synthetic data", "Production pattern" ] }, { "schema": "rapp-agent/1.0", "name": "@aibast-agents-library/warranty-registration", "version": "1.0.0", "display_name": "Warranty and Registration Agent", "description": "Give equipment dealers one place to check warranty coverage, pre-check claims, validate serials and prepare product registrations, so covered repairs move faster and no unit misses its registration window.", "author": "AIBAST", "tags": [ "manufacturing", "warranty", "product-registration", "dealer-service", "aftermarket" ], "category": "manufacturing", "quality_tier": "verified", "requires_env": [], "dependencies": [ "@rapp/basic-agent" ], "_file": "agents/@aibast-agents-library/manufacturing_stacks/warranty_registration_stack/warranty_registration_agent.py", "_size_kb": 23.0, "_lines": 472, "_sha256": "9b198a7bfdd59b5c94c2504fc984d79100b57de2393dce80d4ae69dd90c4a71e", "_install_prefix": "warranty_registration__", "_install_filename": "warranty_registration__9b198a7bfdd5_agent.py", "_stack": "warranty_registration", "_stack_vertical": "manufacturing", "_synthetic_data": true, "_added_at": "2026-10-08T10:30:56-04:00", "_solution": { "advertised_name": "Warranty and Registration Agent", "sharepoint_list_name": "Warranty and Registration Agent", "slot": 64, "executive_summary": "Give equipment dealers one place to check warranty coverage, pre-check claims, validate serials and prepare product registrations, so covered repairs move faster and no unit misses its registration window.", "industries": [ "Manufacturing" ], "personas": [ "Dealer Service Managers", "Warranty Administrators", "Dealer Sales Staff" ], "featured_tools": [ "Microsoft Teams", "Dynamics 365 Field Service", "Microsoft Outlook" ], "agent_requirements": [ "Microsoft Copilot Studio" ], "capabilities": [ "Shows coverage, expirations and pending registrations in one view", "Pre-checks repair coverage and pulls full unit records", "Validates serials and drafts registrations for review", "Drafts extended-coverage offers for units ending soon" ], "outcomes": [ "Fewer rejected warranty claims", "Registrations completed inside the window", "More extensions sold before coverage ends" ], "customer_scenario": [ "Coverage lookups span several disconnected screens", "Claims are rejected for lapsed or parts-only coverage", "Sold units miss their registration window", "Extension opportunities expire unnoticed" ], "slide_title": "Warranty and Registration Agent", "slide_summary": "Gives equipment dealers one place to check warranty coverage and prepare registrations, so covered repairs move faster.", "scenario_name": "A regional equipment dealer", "challenge_intro": "Dealer service teams juggle warranty, catalog and registration systems:", "actions_intro": "The Warranty and Registration Agent gives dealers one conversation:", "outcomes_intro": "Dealers that adopt the agent can expect:", "opportunity_statements": [ "Check coverage before the repair, not after the claim", "Register every sold unit inside its window", "Offer extended coverage before it lapses" ], "module_name": null, "module_description": null, "source_verified": true, "sales_headline": "Know what a repair is covered for before the wrench comes out", "card_pitch": "Give equipment dealers one conversation for warranty coverage, claim pre-checks, serial checks, registration drafts and extended-coverage offers, while claims, registrations and sales stay with the people authorized to submit them.", "why_try": "The demo walks a dealer service manager from a coverage overview to a covered repair, an expired unit, a clean registration draft and a short list of extension offers, each answer computed from the same fixed snapshot.", "customer_challenge_copy": "Dealers look up coverage across separate warranty, catalog and registration screens, miss registration windows on newly sold units, file claims that are rejected for lapsed or parts-only coverage, and let extension opportunities expire unnoticed.", "microsoft_ai_story": "Copilot Studio gives dealers a Teams or Microsoft 365 Copilot front door; approved ERP warranty, product catalog and registration services can be bound with dealer-scoped identity, least-privilege connections and audit, while every claim, registration and offer stays a human-submitted draft.", "business_value": [ "Fewer rejected warranty claims because coverage is checked before the repair.", "No sold unit misses its registration window.", "Extended-coverage offers reach customers before coverage ends." ], "search_terms": [ "Dealer Sales Staff", "Dealer Service Manager", "Dynamics 365 Field Service", "Microsoft Copilot Studio", "Microsoft Outlook", "Microsoft Teams", "Warranty Administrator", "Warranty and Registration Agent", "claim_precheck", "coverage_overview", "extension_offers", "registration_draft", "unit_record", "validate_serial" ], "journey_stage": "Serve", "blueprint_role": "Puts a dealer-facing warranty and registration layer in front of the manufacturer's systems of record, separating pre-checks and drafts from claim decisions and submissions.", "sample_prompts": [ { "label": "Dealer Service Manager: Coverage overview", "prompt": "How does our dealership's warranty coverage look right now?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=warranty-registration-coverage-overview" }, { "label": "Dealer Service Manager: Claim pre-check", "prompt": "A customer brought in welding station WS-25-4471 with a failed wire feeder. Is the repair covered?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=warranty-registration-claim-precheck" }, { "label": "Warranty Administrator: Unit warranty record", "prompt": "Pull the full warranty record for tire changer TC-90-3307.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=warranty-registration-unit-record" }, { "label": "Dealer Sales Staff: Serial check", "prompt": "We just sold welding station WS-25-4520. Check the serial before we register it.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=warranty-registration-validate-serial" }, { "label": "Dealer Sales Staff: Registration draft", "prompt": "Prepare the registration for WS-25-4520, but don't submit anything.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=warranty-registration-registration-draft" }, { "label": "Dealer Sales Staff: Extended-coverage offers", "prompt": "Which units should we offer extended coverage to this month?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=warranty-registration-extension-offers" } ], "architecture": { "business_flow": [ "Dealer service and sales staff", "Microsoft Teams or Microsoft 365 Copilot", "Warranty and Registration Agent", "Approved ERP warranty module and product catalog", "Dealer registration service", "Manufacturer warranty desk (human decision)" ], "capabilities": [ { "operation": "coverage_overview", "name": "Coverage overview", "purpose": "Use when a dealer asks how their warranty coverage looks overall: active, expiring, expired and awaiting registration." }, { "operation": "claim_precheck", "name": "Claim pre-check", "purpose": "Use when a dealer asks whether a repair on a specific serial is covered; returns a pre-check, never files a claim." }, { "operation": "unit_record", "name": "Unit warranty record", "purpose": "Use when someone asks for the full warranty record or claim history of one serial." }, { "operation": "validate_serial", "name": "Serial check", "purpose": "Use when a dealer wants to check a serial (format, catalog, duplicate) before registering a sold unit." }, { "operation": "registration_draft", "name": "Registration draft", "purpose": "Use when a dealer asks to prepare or draft a product registration; the draft stays Not Submitted." }, { "operation": "extension_offers", "name": "Extended-coverage offers", "purpose": "Use when a dealer asks which units to offer extended coverage to; returns offer drafts, never sends or sells." } ], "easy_mode": [ "Verify Power Platform CLI 2.9.3 or newer and authenticate to the approved target environment.", "Initialize a modern cli-copilot project with an approved publisher prefix and preserve the generated identity settings.", "Upload the two packaged synthetic knowledge files and every operation-specific SKILL.md.", "Bind only approved least-privilege production connections after security, privacy, and business-owner review.", "Push the Draft agent, replay every locked persona case, and stop before publish for explicit approval." ], "local_install_prompt": "Install Warranty and Registration Agent from https://raw.githubusercontent.com/microsoft/aibast-agents-library/main/solutions/warranty-registration/deployment.json, verify the expected tool and smoke case, and preserve the synthetic read-only boundary. Do not ask me to open a terminal. Do not connect customer systems or execute external actions.", "copilot_studio_prompt": "Using the Microsoft Copilot Studio plugin, initialize a Draft agent for Warranty and Registration Agent from https://raw.githubusercontent.com/microsoft/aibast-agents-library/main/solutions/warranty-registration/deployment.json. Upload the two synthetic knowledge files and 6 skills, replay every locked case, and Stop before publish. Do not bind production connections without explicit customer approval.", "required_connections": [ "Use only approved, least-privilege production connections; the packaged demonstration uses synthetic knowledge files.", "Approved ERP warranty module (read-only, dealer-scoped)", "Approved product catalog and registration service (draft-only)" ], "manual_commands": [ "pac auth create", "pac copilot init --name \"Warranty and Registration Agent\" --publisher-prefix --authoring-mode cli-copilot --project-dir \"\" --environment \"\"", "pac connection list --environment \"\"", "pac copilot pull --project-dir \"\"", "pac copilot push --project-dir \"\"", "pac copilot publish --bot \"\" --environment \"\"" ], "acceptance_checks": [ "Every exposed operation has one persona-language case and one SKILL.md.", "Exactly two synthetic knowledge files are present.", "Responses state the relevant human approval or review boundary.", "No operation claims an external write, communication, commitment, or protected decision.", "Publishing remains a separate user-approved step." ], "hard_mode": [ "Run pac copilot init and verify settings.mcs.yml before authoring components.", "Author global instructions that preserve the packaged synthetic-only and human-approval boundaries.", "Add the two knowledge files and one skill for every exposed operation.", "Create dedicated connection references only for approved systems and verify least-privilege access.", "Pull before push, inspect every generated file, replay every locked case, and stop before publish." ] }, "curated_copy": true, "source": "field", "origin": "both", "onepager": "64. Warranty and Registration Agent one-pager.pptx", "demo_video": "64-Warranty and Registration Agent.mp4", "has_onepager": true, "has_demo_video": true, "promise_coverage": 100, "is_primary": true, "aliases": [], "package": { "slug": "warranty-registration", "quest_url": "solutions/warranty-registration/quest.html", "manual_tutorial_url": "solutions/warranty-registration/manual-tutorial.html", "export_manifest_url": "solutions/warranty-registration/export-manifest.json" } }, "_demo": { "slug": "warranty-registration", "case_count": 6, "personas": [ "Dealer Sales Staff", "Dealer Service Manager", "Warranty Administrator" ], "prompts": [ "How does our dealership's warranty coverage look right now?", "A customer brought in welding station WS-25-4471 with a failed wire feeder. Is the repair covered?", "Pull the full warranty record for tire changer TC-90-3307." ] }, "_catalog_kind": "solution", "_readiness": [ "Demo proven", "Conversation tested", "Synthetic data", "Production pattern" ] }, { "schema": "rapp-agent/1.0", "name": "@aibast-agents-library/client-health-score", "version": "1.1.1", "display_name": "Client Health Score Agent", "description": "Automate client portfolio health monitoring and planning to improve client relationships, protect revenue, and optimize financial performance.", "author": "AIBAST", "tags": [ "client-health", "NPS", "retention", "professional-services", "churn" ], "category": "professional_services", "quality_tier": "verified", "requires_env": [], "dependencies": [ "@rapp/basic-agent" ], "_file": "agents/@aibast-agents-library/professional_services_stacks/client_health_score_stack/client_health_score_agent.py", "_size_kb": 34.9, "_lines": 741, "_sha256": "6bc19d3b775c7e32d5d8fe080023b939288f26b35f149cc4757987d2e9bb0e7b", "_install_prefix": "client_health_score__", "_install_filename": "client_health_score__6bc19d3b775c_agent.py", "_stack": "client_health_score", "_stack_vertical": "professional_services", "_synthetic_data": true, "_added_at": "2026-03-18T13:23:12-04:00", "_solution": { "advertised_name": "Client Health Score Agent", "sharepoint_list_name": "Client Health Score Agent", "slot": 16, "executive_summary": "Automate client portfolio health monitoring and planning to improve client relationships, protect revenue, and optimize financial performance.", "industries": [ "Professional Services, Consulting" ], "personas": [ "Client Success Leaders", "Account Manager" ], "featured_tools": [ "Dynamics 365", "Microsoft Teams", "Outlook" ], "agent_requirements": [ "Microsoft Copilot Studio" ], "capabilities": [ "Implement real-time health scoring and churn probability modeling", "Segment accounts into critical, at-risk, and healthy categories for prioritization", "Develop tailored retention playbooks, stakeholder mapping, and executive engagement plans" ], "outcomes": [ "Protected revenue by safeguarding high-value relationships through structured turnaround plans", "Identified relationship opportunities and improved client satisfaction", "Enabled proactive intervention and portfolio stabilization, reducing churn risk" ], "customer_scenario": [ "Satisfaction scores were declining, creating renewal uncertainty", "Manual portfolio reviews drained time without providing early warning signals", "Due to time constraints, sellers developed incomplete retention plans leading to limited impact" ], "slide_title": "Client Health Score Agent", "slide_summary": "Automate client portfolio health monitoring and planning to improve client relationships, protect revenue, and optimize financial performance.", "scenario_name": "A professional services firm", "challenge_intro": "A consulting firm faced retention challenges across its client base:", "actions_intro": "The firm leveraged the Client Health Score Agent to:", "outcomes_intro": "With the Agent, the firm:", "opportunity_statements": [ "Perform predictive health scoring to identify risks early and enable proactive retention strategies", "Detect at-risk accounts to prevent churn and protect high-value client relationships", "Automate retention planning to restore satisfaction and strengthen executive trust" ], "module_name": null, "module_description": null, "source_verified": true, "sales_headline": "Turn scattered relationship signals into an approval-ready retention plan", "card_pitch": "Help client success leaders and account managers segment relationship health, spot weakening engagement, prioritize turnaround accounts, and prepare stakeholder-led recovery plans without presenting scenario indicators as certain outcomes.", "why_try": "The demo moves from a portfolio question to the specific relationship signals, owners, evidence, and executive engagement needed for a governed intervention.", "customer_challenge_copy": "A professional services firm has relationship, delivery, financial, and satisfaction signals spread across teams. Manual reviews can surface risk late, overstate the meaning of a score, and produce incomplete retention plans without clear ownership.", "microsoft_ai_story": "Microsoft Copilot Studio provides the portfolio conversation. Dynamics 365 supplies governed account context, Microsoft Teams supports internal coordination, and Outlook supports approved executive engagement.", "business_value": [ "Improves consistency in portfolio prioritization.", "Helps teams connect risk signals with accountable recovery work.", "Strengthens preparation for executive engagement and retention planning." ], "search_terms": [ "client health", "account risk", "churn indicator", "retention plan", "client satisfaction", "executive engagement", "stakeholder map", "Dynamics 365", "Microsoft Teams", "Outlook", "Copilot Studio" ], "journey_stage": "Optimize", "blueprint_role": "Creates a governed relationship-intelligence layer between portfolio signals and account-owner-approved intervention.", "sample_prompts": [ { "label": "Segment the portfolio", "prompt": "Which relationships are healthy, at risk, or critical, and where should my team focus first?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=client-portfolio" }, { "label": "Inspect engagement weakness", "prompt": "What engagement signals are weakening across the portfolio, especially executive contact and escalations?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=client-engagement" }, { "label": "Review satisfaction movement", "prompt": "Which client satisfaction trends are moving the wrong way, and what evidence supports that view?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=client-satisfaction" }, { "label": "Prioritize intervention", "prompt": "Give me the accounts that need intervention now, the risk drivers, and the first recovery actions.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=client-risk" }, { "label": "Prepare turnaround playbooks", "prompt": "Build the stakeholder map and executive engagement plan for each account that needs a turnaround.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=client-retention" } ], "architecture": { "business_flow": [ "Client success leaders and account managers", "Microsoft Copilot or Copilot Studio", "Client Health Score Agent", "Dynamics 365", "Microsoft Teams", "Outlook", "Account-owner-approved intervention" ], "capabilities": [ { "operation": "health_dashboard", "name": "Portfolio health segmentation", "purpose": "Ranks relationship health and clearly labels scenario churn indicators." }, { "operation": "engagement_analysis", "name": "Engagement signal analysis", "purpose": "Connects executive contact, escalations, billing direction, and utilization." }, { "operation": "satisfaction_trend", "name": "Satisfaction trend review", "purpose": "Shows which relationships are improving, stable, or declining." }, { "operation": "at_risk_clients", "name": "At-risk account prioritization", "purpose": "Explains the drivers and first recovery actions for accounts needing intervention." }, { "operation": "retention_plan", "name": "Stakeholder-led retention planning", "purpose": "Creates account-specific ownership, preparation, engagement, and approval gates." } ], "easy_mode": [ "Initialize a modern Copilot Studio project.", "Use the deployment recipe and blueprint to author the client-health agent.", "Upload the synthetic portfolio and retention-playbook knowledge files.", "Bind approved Dynamics 365, Microsoft Teams, and Outlook connections.", "Replay all locked prompts and stop before publishing." ], "local_install_prompt": "You are GitHub Copilot running in Agent mode in VS Code. Install and open the AI BAST Client Health Score Agent from https://raw.githubusercontent.com/microsoft/aibast-agents-library/main/solutions/client-health-score/deployment.json. Verify the source and expected tool, run the smoke prompt, and open the local UI. Do not contact clients, create meetings, change CRM, connect customer systems, or publish. Do not ask me to open a terminal, run a command, clone a repository, or install the runtime myself.", "copilot_studio_prompt": "You are GitHub Copilot running in Agent mode with the Microsoft Copilot Studio plugin. Promote the AI BAST Client Health Score Agent from its deployment recipe. Preserve all operations, uncertainty labels, and account-owner approval gates, upload the packaged knowledge and skills, bind approved Dynamics 365, Microsoft Teams, and Outlook connections, validate every locked prompt, and stop before publish. Stop before publish.", "required_connections": [ "Dynamics 365", "Microsoft Teams", "Outlook" ], "manual_commands": [ "pac auth create", "pac copilot init --name \"Client Health Score Agent\" --publisher-prefix --authoring-mode cli-copilot --project-dir \"\" --environment \"\"", "pac connection list --environment \"\"", "pac copilot pull --project-dir \"\"", "pac copilot push --project-dir \"\"", "pac copilot publish --bot \"\" --environment \"\"" ], "acceptance_checks": [ "Every client-health workflow is represented.", "Scenario indicators are never described as validated predictions or certainty.", "Turnaround accounts have named internal ownership and an approval gate.", "No meeting, message, concession, renewal, or CRM update is executed.", "Publishing remains separately authorized." ], "hard_mode": [ "Create the agent manually.", "Upload both synthetic client-health knowledge files.", "Create one skill for each operation.", "Add approved Dynamics 365, Microsoft Teams, and Outlook connections.", "Replay every locked case and confirm no external side effect occurred." ] }, "curated_copy": true, "source": "catalog", "origin": null, "onepager": "Client Health Score Agent one-pager.pptx", "demo_video": "16-Client Health Score Agent.mp4", "has_onepager": true, "has_demo_video": true, "promise_coverage": 70, "is_primary": true, "aliases": [], "package": { "slug": "client-health-score", "quest_url": "solutions/client-health-score/quest.html", "manual_tutorial_url": "solutions/client-health-score/manual-tutorial.html", "export_manifest_url": "solutions/client-health-score/export-manifest.json" } }, "_demo": { "slug": "client-health-score", "case_count": 8, "personas": [ "Account Manager", "Client Experience Director", "Client Success Leader" ], "prompts": [ "Which relationships are healthy, at risk, or critical, and where should my team focus first?", "What engagement signals are weakening across the portfolio, especially executive contact and escalations?", "Which client satisfaction trends are moving the wrong way, and what evidence supports that view?" ] }, "_catalog_kind": "solution", "_readiness": [ "Demo proven", "Conversation tested", "Synthetic data", "Production pattern" ] }, { "schema": "rapp-agent/1.0", "name": "@aibast-agents-library/contract-risk-review", "version": "1.2.0", "display_name": "Contract Risk Review Agent", "description": "Automate contract review processes to enable faster, lower-risk, and more successful negotiations.", "author": "AIBAST", "tags": [ "contract", "risk", "legal", "compliance", "professional-services" ], "category": "professional_services", "quality_tier": "verified", "requires_env": [], "dependencies": [ "@rapp/basic-agent" ], "_file": "agents/@aibast-agents-library/professional_services_stacks/contract_risk_review_stack/contract_risk_review_agent.py", "_size_kb": 35.7, "_lines": 702, "_sha256": "31642fc92a4f7050b2a0050389d57d6ced4863aaf6e4a086a4e09979d832d9db", "_install_prefix": "contract_risk_review__", "_install_filename": "contract_risk_review__31642fc92a4f_agent.py", "_stack": "contract_risk_review", "_stack_vertical": "professional_services", "_synthetic_data": true, "_added_at": "2026-03-18T13:23:12-04:00", "_solution": { "advertised_name": "Contract Risk Review Agent", "sharepoint_list_name": "Contract Review Agent", "slot": 2, "executive_summary": "Automate contract review processes to enable faster, lower-risk, and more successful negotiations.", "industries": [ "Professional Services, Consulting" ], "personas": [ "Legal Operations", "Attorneys", "Executives" ], "featured_tools": [ "SharePoint", "Microsoft Word" ], "agent_requirements": [ "Microsoft Copilot Studio" ], "capabilities": [ "Analyze the agreement in seconds, highlighting key issues and risks", "Identify gaps in liability protection, IP ownership, and payment terms compared to industry standards", "Generate a prioritized list of amendments and negotiation strategies" ], "outcomes": [ "Reduced liability exposure and safeguarded intellectual property and proprietary assets", "Improved financial outcomes by optimizing payment terms", "Lowered legal costs and accelerated deal closure" ], "customer_scenario": [ "Initial concerns about liability, IP assignment, and payment terms", "Limited time to analyze the MSA and identify gaps, risks, and prioritized amendments", "Heavy reliance on manual analysis and cross-referencing processes" ], "slide_title": "Contract Risk Review Agent", "slide_summary": "Automate contract review processes to enable faster, lower-risk, and more successful negotiations.", "scenario_name": "Professional services firm", "challenge_intro": "The firm received a lengthy master services agreement with:", "actions_intro": "The professional services firm used the Contract Risk Review Agent to:", "outcomes_intro": "With the Agent, the firm:", "opportunity_statements": [ "Perform comprehensive contract analysis to accelerate reviews and highlight critical risks", "Evaluate liability exposure and IP provisions to identify gaps and ensure complete financial protections", "Recommend amendments and negotiation strategies to strengthen contract positions and limit legal costs" ], "module_name": null, "module_description": null, "source_verified": true, "sales_headline": "Move the right contract issues to counsel before they slow the deal", "card_pitch": "Give legal operations and commercial counsel a shared view of agreement priority, clause risk, policy gaps, and negotiation positions without turning an AI review into a legal conclusion.", "why_try": "A single plain-language request surfaces the agreements, clauses, evidence gaps, and counsel decisions that deserve attention next.", "customer_challenge_copy": "A professional services firm receives complex agreements under tight deal timelines. Liability, intellectual property, payment, termination, and operational terms are difficult to compare consistently, while missing evidence can be mistaken for acceptable language.", "microsoft_ai_story": "Microsoft Copilot Studio provides the conversational review experience. SharePoint supplies governed agreement evidence, and Microsoft Word supports the counsel-controlled redline and negotiation workflow.", "business_value": [ "Improves consistency in contract intake and prioritization.", "Helps counsel focus on material clause and policy gaps.", "Creates a clearer handoff from review findings to negotiation strategy." ], "search_terms": [ "contract review", "legal operations", "liability clause", "IP ownership", "payment terms", "contract policy", "negotiation brief", "SharePoint contracts", "Microsoft Word", "Copilot Studio", "professional services" ], "journey_stage": "Govern", "blueprint_role": "Establishes a governed review layer between agreement intake and counsel-controlled negotiation.", "sample_prompts": [ { "label": "Prioritize the contract queue", "prompt": "Which agreements should I put in front of counsel first, and what is driving the priority?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=contract-priority" }, { "label": "Inspect the MSA clauses", "prompt": "Walk me through the liability, IP ownership, and payment-term language that needs attention in the MSA.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=contract-clauses" }, { "label": "Find policy and evidence gaps", "prompt": "Compare the available contract evidence with our internal policy and tell me where the file is incomplete.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=contract-policy" }, { "label": "Prepare negotiation positions", "prompt": "Prepare the negotiation positions, fallbacks, and escalation points for the highest-risk agreements.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=contract-negotiation" } ], "architecture": { "business_flow": [ "Legal operations and commercial counsel", "Microsoft Copilot or Copilot Studio", "Contract Risk Review Agent", "SharePoint", "Microsoft Word", "Authorized legal review" ], "capabilities": [ { "operation": "risk_scan", "name": "Portfolio risk triage", "purpose": "Ranks the agreement queue and renewal context for counsel attention." }, { "operation": "clause_analysis", "name": "Clause-level issue analysis", "purpose": "Explains documented liability, intellectual-property, payment, and operational term findings." }, { "operation": "compliance_check", "name": "Internal-policy screen", "purpose": "Compares available evidence with policy and preserves a review-required state when evidence is absent." }, { "operation": "renegotiation_brief", "name": "Negotiation brief preparation", "purpose": "Separates priority amendments, fallback positions, and counsel escalation points." } ], "easy_mode": [ "Initialize a modern Copilot Studio project in the target environment.", "Provide this blueprint and the deployment recipe to the Copilot Studio Architect workflow.", "Upload the packaged synthetic agreement and policy knowledge files.", "Bind approved SharePoint and Microsoft Word connection references.", "Push the draft, replay every locked prompt, and stop before publishing." ], "local_install_prompt": "You are GitHub Copilot running in Agent mode in VS Code. Install and open the AI BAST Contract Risk Review Agent from https://raw.githubusercontent.com/microsoft/aibast-agents-library/main/solutions/contract-risk-review/deployment.json. Own the setup, verify the source against the registry, confirm the expected tool in Brainstem health, run the smoke prompt, and open the local UI. Do not connect customer agreements, provide legal advice, publish, or silently substitute another agent. Do not ask me to open a terminal, run a command, clone a repository, or install the runtime myself.", "copilot_studio_prompt": "You are GitHub Copilot running in Agent mode with the Microsoft Copilot Studio plugin. Promote the AI BAST Contract Risk Review Agent from its deployment recipe. Ask only for the target environment and publisher prefix. Preserve every advertised operation, upload the packaged knowledge and skills, bind approved SharePoint and Microsoft Word connections, validate all locked prompts, and stop before publish with unresolved legal-review and connection work visible. Stop before publish.", "required_connections": [ "SharePoint", "Microsoft Word" ], "manual_commands": [ "pac auth create", "pac copilot init --name \"Contract Risk Review Agent\" --publisher-prefix --authoring-mode cli-copilot --project-dir \"\" --environment \"\"", "pac connection list --environment \"\"", "pac copilot pull --project-dir \"\"", "pac copilot push --project-dir \"\"", "pac copilot publish --bot \"\" --environment \"\"" ], "acceptance_checks": [ "Every advertised contract capability is represented by instructions, a skill, or a bound tool.", "Missing clause evidence returns review required rather than pass.", "Outputs preserve the legal-advice and no-side-effect boundaries.", "All locked prompts route to the expected operation and synthetic evidence.", "Publishing remains a separate authorized step." ], "hard_mode": [ "Create the agent manually in Copilot Studio.", "Upload both synthetic knowledge files.", "Create one skill for each operation.", "Add approved SharePoint and Microsoft Word connections.", "Replay every locked case and verify counsel approval gates before publishing." ] }, "curated_copy": true, "source": "catalog", "origin": null, "onepager": "Contract Risk Review Agent one-pager.pptx", "demo_video": "#2-Contract Risk Review Agent.mp4", "has_onepager": true, "has_demo_video": true, "promise_coverage": 65, "is_primary": true, "aliases": [], "package": { "slug": "contract-risk-review", "quest_url": "solutions/contract-risk-review/quest.html", "manual_tutorial_url": "solutions/contract-risk-review/manual-tutorial.html", "export_manifest_url": "solutions/contract-risk-review/export-manifest.json" } }, "_demo": { "slug": "contract-risk-review", "case_count": 9, "personas": [ "Commercial Attorney", "General Counsel", "Legal Operations Director", "Legal Operations Manager" ], "prompts": [ "Which agreements should I put in front of counsel first, and what is driving the priority?", "Walk me through the liability, IP ownership, and payment-term language that needs attention in the MSA.", "Compare the available contract evidence with our internal policy and tell me where the file is incomplete." ] }, "_catalog_kind": "solution", "_readiness": [ "Demo proven", "Conversation tested", "Synthetic data", "Production pattern" ] }, { "schema": "rapp-agent/1.0", "name": "@aibast-agents-library/resource-utilization", "version": "1.1.0", "display_name": "Resource Utilization Agent", "description": "Provide intelligent resource analysis and recommendations to maximize billable utilization and reduce costs.", "author": "AIBAST", "tags": [ "utilization", "staffing", "capacity", "bench", "professional-services" ], "category": "professional_services", "quality_tier": "verified", "requires_env": [], "dependencies": [ "@rapp/basic-agent" ], "_file": "agents/@aibast-agents-library/professional_services_stacks/resource_utilization_stack/resource_utilization_agent.py", "_size_kb": 34.2, "_lines": 639, "_sha256": "308622fa5077130c0b61475f8a3b51895f099f20d28333232ef4998d10a1b031", "_install_prefix": "resource_utilization__", "_install_filename": "resource_utilization__308622fa5077_agent.py", "_stack": "resource_utilization", "_stack_vertical": "professional_services", "_synthetic_data": true, "_added_at": "2026-03-18T13:23:12-04:00", "_solution": { "advertised_name": "Resource Utilization Agent", "sharepoint_list_name": "Resource Utilization Agent", "slot": 9, "executive_summary": "Provide intelligent resource analysis and recommendations to maximize billable utilization and reduce costs.", "industries": [ "Professional Services, Consulting" ], "personas": [ "Operations Leader", "Resource Manager", "Finance Director" ], "featured_tools": [ "Dynamics 365", "Microsoft Teams" ], "agent_requirements": [ "Microsoft Copilot Studio" ], "capabilities": [ "Implement real-time utilization tracking and capacity analysis", "Give data-driven recommendations for balancing skills and capacity", "Match consultants to projects and high-probability opportunities", "Identify consultants for upskilling pathways and innovation initiatives" ], "outcomes": [ "Reduced bench costs and improved billable utilization beyond targets", "Accelerated project staffing while maintaining quality standards", "Enabled proactive workforce planning and strategic capability building" ], "customer_scenario": [ "Manual resource planning slowed response to urgent pipeline needs", "Leadership lacked visibility into skill gaps and attrition risks", "Forecast confidence was at risk due to unclear deployment priorities" ], "slide_title": "Resource Utilization Agent", "slide_summary": "Provide intelligent resource analysis and recommendations to maximize billable utilization and reduce costs.", "scenario_name": "A professional services firm", "challenge_intro": "A consulting firm faced challenges in improving utilization to combat escalating bench costs:", "actions_intro": "The consulting firm leveraged the Resource Utilization Agent to:", "outcomes_intro": "With the agent\u2019s help, the firm:", "opportunity_statements": [ "Analyze workforce skills and availability to identify deployment gaps and opportunities", "Match consultants to pipeline projects and innovation initiatives for immediate revenue impact", "Model upskilling ROI and implement strategic workforce planning to align with market demand" ], "module_name": null, "module_description": null, "source_verified": true, "sales_headline": "Connect available skills to demand without turning a scenario into a staffing promise", "card_pitch": "Give operations, resource, and finance leaders a shared view of utilization, capacity, bench skills, pipeline matches, and capability-building options while keeping assignments and financial commitments under human control.", "why_try": "A resource manager can ask who is available, see where skills align with demand, and identify the people who need upskilling or an internal innovation path.", "customer_challenge_copy": "Professional services leaders need to balance utilization, urgent demand, skill fit, bench cost, and capability development. Manual planning can double-count people, overstate uncertain pipeline, and leave unmatched talent without a deliberate path.", "microsoft_ai_story": "Microsoft Copilot Studio provides the planning conversation. Dynamics 365 supplies project and resource context, while Microsoft Teams supports governed staffing and development decisions.", "business_value": [ "Improves shared visibility into skills, availability, and capacity.", "Accelerates evidence-based staffing conversations.", "Supports deliberate upskilling and capability-building decisions." ], "search_terms": [ "resource utilization", "bench analysis", "capacity forecast", "staffing match", "skills inventory", "upskilling ROI", "workforce planning", "Dynamics 365", "Microsoft Teams", "Copilot Studio", "professional services" ], "journey_stage": "Optimize", "blueprint_role": "Adds a governed capacity and capability-planning layer across delivery demand, available talent, and workforce investment.", "sample_prompts": [ { "label": "See utilization and availability", "prompt": "Where are we under target today, and which parts of the workforce are sitting available?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=resource-utilization" }, { "label": "Compare capacity with demand", "prompt": "What capacity is coming free, and how does it line up with the opportunities expected over the next quarter?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=resource-capacity" }, { "label": "Understand the bench", "prompt": "Show me who is on the bench, the skills we are carrying, and the cost exposure we need to address.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=resource-bench" }, { "label": "Match people to pipeline", "prompt": "Which available consultants fit the strongest pipeline needs, and who still needs another path?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=resource-staffing" }, { "label": "Plan capability building", "prompt": "Give me an upskilling and internal-innovation plan for the people we cannot place directly, including the business case.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=resource-workforce-plan" } ], "architecture": { "business_flow": [ "Operations, resource, and finance leaders", "Microsoft Copilot or Copilot Studio", "Resource Utilization Agent", "Dynamics 365", "Microsoft Teams", "Authorized staffing and development review" ], "capabilities": [ { "operation": "utilization_dashboard", "name": "Utilization and availability view", "purpose": "Shows current portfolio and level performance against planning targets." }, { "operation": "capacity_forecast", "name": "Capacity and demand forecast", "purpose": "Compares upcoming availability with weighted opportunity needs." }, { "operation": "bench_analysis", "name": "Bench skill and cost analysis", "purpose": "Surfaces available talent, skills, and planning cost." }, { "operation": "staffing_recommendation", "name": "Pipeline staffing matches", "purpose": "Matches only qualified available people and keeps unmatched resources visible." }, { "operation": "workforce_plan", "name": "Upskilling and innovation planning", "purpose": "Models capability-building options without approving assignments or benefits." } ], "easy_mode": [ "Initialize a modern Copilot Studio project.", "Use the deployment recipe and blueprint to author the planning agent.", "Upload the synthetic roster and pipeline-rule knowledge files.", "Bind approved Dynamics 365 and Microsoft Teams connections.", "Replay all locked prompts and stop before publishing." ], "local_install_prompt": "You are GitHub Copilot running in Agent mode in VS Code. Install and open the AI BAST Resource Utilization Agent from https://raw.githubusercontent.com/microsoft/aibast-agents-library/main/solutions/resource-utilization/deployment.json. Verify the source and expected tool, run the smoke prompt, and open the local UI. Do not assign people, change employment records, book revenue, connect customer systems, or publish. Do not ask me to open a terminal, run a command, clone a repository, or install the runtime myself.", "copilot_studio_prompt": "You are GitHub Copilot running in Agent mode with the Microsoft Copilot Studio plugin. Promote the AI BAST Resource Utilization Agent from its deployment recipe. Preserve every operation and approval boundary, upload the packaged knowledge and skills, bind approved Dynamics 365 and Microsoft Teams connections, validate all locked prompts, and stop before publish. Stop before publish.", "required_connections": [ "Dynamics 365", "Microsoft Teams" ], "manual_commands": [ "pac auth create", "pac copilot init --name \"Resource Utilization Agent\" --publisher-prefix --authoring-mode cli-copilot --project-dir \"\" --environment \"\"", "pac connection list --environment \"\"", "pac copilot pull --project-dir \"\"", "pac copilot push --project-dir \"\"", "pac copilot publish --bot \"\" --environment \"\"" ], "acceptance_checks": [ "All resource workflows are represented.", "Matches require both skill and level.", "People are counted once in projections.", "Pipeline and value scenarios remain uncommitted.", "Assignments, training, and publishing remain separately authorized." ], "hard_mode": [ "Create the agent manually.", "Upload both synthetic workforce knowledge files.", "Create one skill for each operation.", "Add approved Dynamics 365 and Microsoft Teams connections.", "Replay every locked case and verify no staffing side effect occurred." ] }, "curated_copy": true, "source": "catalog", "origin": null, "onepager": "Resource Utilization Agent one-pager.pptx", "demo_video": "17-Resource Utilization Agent.mp4", "has_onepager": true, "has_demo_video": true, "promise_coverage": 70, "is_primary": true, "aliases": [], "package": { "slug": "resource-utilization", "quest_url": "solutions/resource-utilization/quest.html", "manual_tutorial_url": "solutions/resource-utilization/manual-tutorial.html", "export_manifest_url": "solutions/resource-utilization/export-manifest.json" } }, "_demo": { "slug": "resource-utilization", "case_count": 8, "personas": [ "Finance Director", "Operations Leader", "Resource Manager" ], "prompts": [ "Where are we under target today, and which parts of the workforce are sitting available?", "What capacity is coming free, and how does it line up with the opportunities expected over the next quarter?", "Show me who is on the bench, the skills we are carrying, and the cost exposure we need to address." ] }, "_catalog_kind": "solution", "_readiness": [ "Demo proven", "Conversation tested", "Synthetic data", "Production pattern" ] }, { "schema": "rapp-agent/1.0", "name": "@aibast-agents-library/solution-design-review", "version": "1.0.0", "display_name": "Solution Design Review Agent", "description": "Help architects draft and self-check a solution design document before the architecture review board: intake of the template, discovery notes and initiative brief, an architecture vision, the current-state baseline, the non-functional requirements matrix with owners and gaps, the target design with its inventory changes, and a readiness check with a confidence score and severity-ranked gaps. The agent is read-only; it never submits to the review board, changes an inventory record, or approves a design.", "author": "AIBAST", "tags": [ "architecture", "solution-design", "design-review", "non-functional-requirements", "professional-services" ], "category": "professional_services", "quality_tier": "verified", "requires_env": [], "dependencies": [ "@rapp/basic-agent" ], "_file": "agents/@aibast-agents-library/professional_services_stacks/solution_design_review_stack/solution_design_review_agent.py", "_size_kb": 20.7, "_lines": 414, "_sha256": "6ec7dc6dc27b510a2cfee8272062b8a47f4e51978d46d66dab961df18eed0482", "_install_prefix": "solution_design_review__", "_install_filename": "solution_design_review__6ec7dc6dc27b_agent.py", "_stack": "solution_design_review", "_stack_vertical": "professional_services", "_synthetic_data": true, "_added_at": "2026-10-08T10:30:56-04:00", "_solution": { "advertised_name": "Solution Design Review Agent", "sharepoint_list_name": "Solution Design Review Agent", "slot": 56, "executive_summary": "Help architects draft and self-check a solution design document before the architecture review board: intake of the template, discovery notes and initiative brief, an architecture vision, the current-state baseline, the non-functional requirements matrix with owners and gaps, the target design with its inventory changes, and a readiness check with a confidence score and severity-ranked gaps. The agent is read-only; it never submits to the review board, changes an inventory record, or approves a design.", "industries": [ "Professional Services", "Cross-Industry" ], "personas": [ "Solution Architect", "Enterprise Architecture Lead", "Review Board Coordinator" ], "featured_tools": [ "Microsoft Teams", "Microsoft Word" ], "agent_requirements": [ "Microsoft Copilot Studio" ], "capabilities": [ "Plan the design from the template, notes and brief in one step", "Draft the vision, baseline and target design with inventory changes", "Build the requirements matrix with owners, gaps and variances", "Run a readiness check with a confidence score before each review" ], "outcomes": [ "Produced first drafts of design sections in hours instead of days", "Arrived at review with owners named and variances answered", "Reduced designs sent back for avoidable checklist gaps" ], "customer_scenario": [ "Architects spent days copying notes into a long template", "Incomplete inventories let end-of-support systems slip through", "Non-functional requirements lacked owners until review", "Designs were sent back for gaps a checklist could catch" ], "slide_title": "Solution Design Review Agent", "slide_summary": "Helps architects draft and self-check a solution design before the review board, with a confidence score and ranked gaps. It never approves a design.", "scenario_name": "A national distributor", "challenge_intro": "A national distributor's architects were preparing designs for the review board:", "actions_intro": "The architecture team used the Solution Design Review Agent to:", "outcomes_intro": "With the agent's help, the team:", "opportunity_statements": [ "Turn discovery notes and an initiative brief into a structured design draft", "Assemble the current state and target design with every inventory change listed", "Map non-functional requirements to owners and surface gaps early", "Self-check readiness before review with a confidence score and ranked gaps" ], "module_name": null, "module_description": null, "source_verified": true, "sales_headline": "Arrive at the architecture review board ready, not revised", "card_pitch": "Give architects a drafting and self-check partner for solution designs: vision, baseline, requirements matrix, target design and a readiness score, while the architect owns every section and the board owns approval.", "why_try": "The demo takes one fictional initiative from intake to a readiness verdict in six questions, with every gap traced to evidence.", "customer_challenge_copy": "Design documents take days to assemble, requirement owners are missing until the review, and designs bounce back for gaps a checklist could have caught.", "microsoft_ai_story": "Copilot Studio hosts the design conversation in Microsoft Teams; approved document libraries, requirement repositories and the component inventory can be bound later with least-privilege, read-only access.", "business_value": [ "Cuts the time to a complete first draft of a solution design.", "Surfaces owner gaps and variances before the review board does.", "Keeps design authority with the architect and approval with the board." ], "search_terms": [ "Enterprise Architecture Lead", "Microsoft Copilot Studio", "Microsoft Teams", "Review Board Coordinator", "Solution Architect", "Solution Design Review Agent", "architecture review board", "architecture_vision", "current_state", "design readiness", "design_intake", "non-functional requirements", "readiness_check", "requirements_matrix", "solution design document", "target_design" ], "journey_stage": "Empower", "blueprint_role": "Separates drafting and self-checking from protected governance actions such as board submission, inventory registration and design approval.", "sample_prompts": [ { "label": "Solution Architect: Design intake and plan", "prompt": "I have the design template, the discovery notes and the initiative brief for the delivery notifications project. Walk me through where to start.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=solution-design-review-design-intake" }, { "label": "Solution Architect: Architecture vision draft", "prompt": "Draft the vision: scope, stakeholders and which architecture domains are impacted.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=solution-design-review-architecture-vision" }, { "label": "Solution Architect: Current-state baseline", "prompt": "What does the current state look like for the systems in scope?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=solution-design-review-current-state" }, { "label": "Enterprise Architecture Lead: Non-functional requirements matrix", "prompt": "Which non-functional requirements apply, and who owns each one?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=solution-design-review-requirements-matrix" }, { "label": "Solution Architect: Target design and inventory delta", "prompt": "Draft the target design and list what changes in the inventory.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=solution-design-review-target-design" }, { "label": "Review Board Coordinator: Review-board readiness check", "prompt": "Is this design ready for the review board?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=solution-design-review-readiness-check" } ], "architecture": { "business_flow": [ "Solution architects", "Microsoft Teams or Microsoft 365 Copilot", "Solution Design Review Agent", "Approved design template and document library", "Component inventory and requirement repository (read-only seams)", "Architecture review board" ], "capabilities": [ { "operation": "design_intake", "name": "Design intake and plan", "purpose": "Use when an architect hands over the design template, discovery notes and initiative brief and asks where to start." }, { "operation": "architecture_vision", "name": "Architecture vision draft", "purpose": "Use when an architect asks for the vision: business driver, scope, stakeholders, impacted domains, constraints and review workgroups." }, { "operation": "current_state", "name": "Current-state baseline", "purpose": "Use when an architect asks what the current state looks like for the systems or components in scope." }, { "operation": "requirements_matrix", "name": "Non-functional requirements matrix", "purpose": "Use when someone asks which non-functional requirements apply, who owns each one, and where the gaps and variances are." }, { "operation": "target_design", "name": "Target design and inventory delta", "purpose": "Use when an architect asks for the target design and what changes in the component inventory." }, { "operation": "readiness_check", "name": "Review-board readiness check", "purpose": "Use when someone asks whether the design is ready for the review board, its confidence score or its gaps." } ], "easy_mode": [ "Verify Power Platform CLI 2.9.3 or newer and authenticate to the approved target environment.", "Initialize a modern cli-copilot project with an approved publisher prefix and preserve the generated identity settings.", "Upload the two packaged synthetic knowledge files and every operation-specific SKILL.md.", "Bind only approved least-privilege production connections after security, privacy, and business-owner review.", "Push the Draft agent, replay every locked persona case, and stop before publish for explicit approval." ], "local_install_prompt": "Install Solution Design Review Agent from https://raw.githubusercontent.com/microsoft/aibast-agents-library/main/solutions/solution-design-review/deployment.json, verify the expected tool and smoke case, and preserve the synthetic read-only boundary. Do not ask me to open a terminal. Do not connect customer systems or execute external actions.", "copilot_studio_prompt": "Using the Microsoft Copilot Studio plugin, initialize a Draft agent for Solution Design Review Agent from https://raw.githubusercontent.com/microsoft/aibast-agents-library/main/solutions/solution-design-review/deployment.json. Upload the two synthetic knowledge files and 6 skills, replay every locked case, and Stop before publish. Do not bind production connections without explicit customer approval.", "required_connections": [ "Use only approved, read-only document, requirement and inventory connections; the packaged demonstration uses synthetic knowledge files." ], "manual_commands": [ "pac auth create", "pac copilot init --name \"Solution Design Review Agent\" --publisher-prefix --authoring-mode cli-copilot --project-dir \"\" --environment \"\"", "pac connection list --environment \"\"", "pac copilot pull --project-dir \"\"", "pac copilot push --project-dir \"\"", "pac copilot publish --bot \"\" --environment \"\"" ], "acceptance_checks": [ "Every exposed operation has one persona-language case and one SKILL.md.", "Exactly two synthetic knowledge files are present.", "Responses state the relevant human approval or review boundary.", "No operation claims an external write, communication, commitment, or protected decision.", "Publishing remains a separate user-approved step." ], "hard_mode": [ "Run pac copilot init and verify settings.mcs.yml before authoring components.", "Author global instructions that preserve the packaged synthetic-only and human-approval boundaries.", "Add the two knowledge files and one skill for every exposed operation.", "Create dedicated connection references only for approved systems and verify least-privilege access.", "Pull before push, inspect every generated file, replay every locked case, and stop before publish." ] }, "curated_copy": true, "source": "field", "origin": "both", "onepager": "56. Solution Design Review Agent one-pager.pptx", "demo_video": "56-Solution Design Review Agent.mp4", "has_onepager": true, "has_demo_video": true, "promise_coverage": 100, "is_primary": true, "aliases": [], "package": { "slug": "solution-design-review", "quest_url": "solutions/solution-design-review/quest.html", "manual_tutorial_url": "solutions/solution-design-review/manual-tutorial.html", "export_manifest_url": "solutions/solution-design-review/export-manifest.json" } }, "_demo": { "slug": "solution-design-review", "case_count": 6, "personas": [ "Enterprise Architecture Lead", "Review Board Coordinator", "Solution Architect" ], "prompts": [ "I have the design template, the discovery notes and the initiative brief for the delivery notifications project. Walk me through where to start.", "Draft the vision: scope, stakeholders and which architecture domains are impacted.", "What does the current state look like for the systems in scope?" ] }, "_catalog_kind": "solution", "_readiness": [ "Demo proven", "Conversation tested", "Synthetic data", "Production pattern" ] }, { "schema": "rapp-agent/1.0", "name": "@aibast-agents-library/staffing-territory-planning", "version": "1.0.0", "display_name": "Staffing Territory Planning Agent", "description": "Help staffing sales leaders size their territory, find where demand outruns recruiter coverage, read the competition and labor trends, and walk into the quarter with a draft plan grounded in the numbers.", "author": "AIBAST", "tags": [ "professional-services", "staffing", "territory-planning", "market-sizing", "sales-planning" ], "category": "professional_services", "quality_tier": "verified", "requires_env": [], "dependencies": [ "@rapp/basic-agent" ], "_file": "agents/@aibast-agents-library/professional_services_stacks/staffing_territory_planning_stack/staffing_territory_planning_agent.py", "_size_kb": 18.7, "_lines": 373, "_sha256": "b60864abd74fe4ec9f498528c18b4b6dada06c3aa805de22246dee6fa0fe6a84", "_install_prefix": "staffing_territory_planning__", "_install_filename": "staffing_territory_planning__b60864abd74f_agent.py", "_stack": "staffing_territory_planning", "_stack_vertical": "professional_services", "_synthetic_data": true, "_added_at": "2026-10-08T10:30:56-04:00", "_solution": { "advertised_name": "Staffing Territory Planning Agent", "sharepoint_list_name": "Staffing Territory Planning Agent", "slot": 65, "executive_summary": "Help staffing sales leaders size their territory, find where demand outruns recruiter coverage, read the competition and labor trends, and walk into the quarter with a draft plan grounded in the numbers.", "industries": [ "Professional Services, Consulting" ], "personas": [ "Regional Sales Directors", "Branch Managers", "Sales Operations Analysts" ], "featured_tools": [ "Microsoft Teams", "Dynamics 365 Sales", "Microsoft Excel" ], "agent_requirements": [ "Microsoft Copilot Studio" ], "capabilities": [ "Summarizes the territory and sizes each metro's market", "Finds coverage gaps and ranks them by open spend", "Shows competitor share and labor trends by line", "Drafts the quarterly plan for leadership review" ], "outcomes": [ "Faster, consistent territory reviews", "Recruiter capacity aimed at the biggest gaps", "Plans leadership can act on sooner" ], "customer_scenario": [ "Territory reviews are assembled by hand", "Coverage gaps are invisible until revenue misses", "Competitive position is anecdotal", "Plans are slow and inconsistent across regions" ], "slide_title": "Staffing Territory Planning Agent", "slide_summary": "Helps staffing sales leaders size their territory, find where demand outruns recruiter coverage, and walk into the quarter with a draft plan.", "scenario_name": "A regional staffing firm", "challenge_intro": "Staffing sales leaders plan territories from scattered exports and reports:", "actions_intro": "The Staffing Territory Planning Agent gives sales leaders a planning partner:", "outcomes_intro": "Sales organizations that adopt the agent can expect:", "opportunity_statements": [ "See the whole territory in one view", "Aim recruiters where demand outruns coverage", "Walk into the quarter with a draft plan" ], "module_name": null, "module_description": null, "source_verified": true, "sales_headline": "Plan the territory from the numbers, not from memory", "card_pitch": "Give staffing sales leaders a territory snapshot, market sizing, coverage gaps, competitor position, labor trends and a draft quarterly plan from one consistent dataset, while headcount, quota and account decisions stay with leadership.", "why_try": "The demo takes a regional sales director from a territory snapshot to a drafted quarterly plan in six questions, with every figure in the plan traceable to the earlier answers.", "customer_challenge_copy": "Staffing sales leaders assemble territory reviews by hand from CRM exports, placement history and labor-market reports, so planning is slow, inconsistent between regions and rarely shows where demand outruns recruiter coverage.", "microsoft_ai_story": "Copilot Studio gives sales leaders a Teams or Microsoft 365 Copilot planning assistant; approved CRM, placement and labor-market sources can be bound with role-scoped access, while the agent returns analysis and a draft plan and never changes headcount, quotas or accounts.", "business_value": [ "Territory reviews prepared in minutes instead of days.", "Recruiter capacity aimed where demand outruns coverage.", "One consistent planning view across regions and quarters." ], "search_terms": [ "Branch Manager", "Dynamics 365 Sales", "Microsoft Copilot Studio", "Microsoft Excel", "Microsoft Teams", "Regional Sales Director", "Sales Operations Analyst", "Staffing Territory Planning Agent", "competitive_landscape", "coverage_gaps", "labor_trends", "market_sizing", "territory_plan", "territory_snapshot" ], "journey_stage": "Plan and evaluate", "blueprint_role": "Adds a planning layer over CRM and market data that turns territory numbers into a reviewable draft plan, leaving headcount, quota and account decisions to sales leadership.", "sample_prompts": [ { "label": "Regional Sales Director: Territory snapshot", "prompt": "Give me a snapshot of my North Valley territory.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=staffing-territory-planning-territory-snapshot" }, { "label": "Regional Sales Director: Market sizing", "prompt": "How big is the staffing market in Granite Ridge?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=staffing-territory-planning-market-sizing" }, { "label": "Sales Operations Analyst: Coverage gaps", "prompt": "Where are our biggest coverage gaps?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=staffing-territory-planning-coverage-gaps" }, { "label": "Branch Manager: Competitive landscape", "prompt": "Who are we up against in each metro?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=staffing-territory-planning-competitive-landscape" }, { "label": "Regional Sales Director: Labor trends", "prompt": "What labor market trends should I plan around?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=staffing-territory-planning-labor-trends" }, { "label": "Regional Sales Director: Draft territory plan", "prompt": "Draft my quarterly territory plan.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=staffing-territory-planning-territory-plan" } ], "architecture": { "business_flow": [ "Regional sales leaders and branch managers", "Microsoft Teams or Microsoft 365 Copilot", "Staffing Territory Planning Agent", "Approved CRM and placement history", "Approved labor-market data feed", "Sales leadership planning review (human decision)" ], "capabilities": [ { "operation": "territory_snapshot", "name": "Territory snapshot", "purpose": "Use when a sales leader asks for a territory snapshot or how the territory looks." }, { "operation": "market_sizing", "name": "Market sizing", "purpose": "Use when someone asks how big the market is, for the territory or one metro, by service line." }, { "operation": "coverage_gaps", "name": "Coverage gaps", "purpose": "Use when a leader asks where the coverage gaps or under-served markets are." }, { "operation": "competitive_landscape", "name": "Competitive landscape", "purpose": "Use when someone asks about competitors, market share leaders or our rank by metro." }, { "operation": "labor_trends", "name": "Labor trends", "purpose": "Use when someone asks about labor market, hiring or wage trends to plan around." }, { "operation": "territory_plan", "name": "Draft territory plan", "purpose": "Use when a leader asks to draft the quarterly or territory plan; the plan stays a draft for leadership." } ], "easy_mode": [ "Verify Power Platform CLI 2.9.3 or newer and authenticate to the approved target environment.", "Initialize a modern cli-copilot project with an approved publisher prefix and preserve the generated identity settings.", "Upload the two packaged synthetic knowledge files and every operation-specific SKILL.md.", "Bind only approved least-privilege production connections after security, privacy, and business-owner review.", "Push the Draft agent, replay every locked persona case, and stop before publish for explicit approval." ], "local_install_prompt": "Install Staffing Territory Planning Agent from https://raw.githubusercontent.com/microsoft/aibast-agents-library/main/solutions/staffing-territory-planning/deployment.json, verify the expected tool and smoke case, and preserve the synthetic read-only boundary. Do not ask me to open a terminal. Do not connect customer systems or execute external actions.", "copilot_studio_prompt": "Using the Microsoft Copilot Studio plugin, initialize a Draft agent for Staffing Territory Planning Agent from https://raw.githubusercontent.com/microsoft/aibast-agents-library/main/solutions/staffing-territory-planning/deployment.json. Upload the two synthetic knowledge files and 6 skills, replay every locked case, and Stop before publish. Do not bind production connections without explicit customer approval.", "required_connections": [ "Use only approved, least-privilege production connections; the packaged demonstration uses synthetic knowledge files.", "Approved CRM account and placement data (read-only, role-scoped)", "Approved labor-market data feed (read-only)" ], "manual_commands": [ "pac auth create", "pac copilot init --name \"Staffing Territory Planning Agent\" --publisher-prefix --authoring-mode cli-copilot --project-dir \"\" --environment \"\"", "pac connection list --environment \"\"", "pac copilot pull --project-dir \"\"", "pac copilot push --project-dir \"\"", "pac copilot publish --bot \"\" --environment \"\"" ], "acceptance_checks": [ "Every exposed operation has one persona-language case and one SKILL.md.", "Exactly two synthetic knowledge files are present.", "Responses state the relevant human approval or review boundary.", "No operation claims an external write, communication, commitment, or protected decision.", "Publishing remains a separate user-approved step." ], "hard_mode": [ "Run pac copilot init and verify settings.mcs.yml before authoring components.", "Author global instructions that preserve the packaged synthetic-only and human-approval boundaries.", "Add the two knowledge files and one skill for every exposed operation.", "Create dedicated connection references only for approved systems and verify least-privilege access.", "Pull before push, inspect every generated file, replay every locked case, and stop before publish." ] }, "curated_copy": true, "source": "field", "origin": "both", "onepager": "65. Staffing Territory Planning Agent one-pager.pptx", "demo_video": "65-Staffing Territory Planning Agent.mp4", "has_onepager": true, "has_demo_video": true, "promise_coverage": 100, "is_primary": true, "aliases": [], "package": { "slug": "staffing-territory-planning", "quest_url": "solutions/staffing-territory-planning/quest.html", "manual_tutorial_url": "solutions/staffing-territory-planning/manual-tutorial.html", "export_manifest_url": "solutions/staffing-territory-planning/export-manifest.json" } }, "_demo": { "slug": "staffing-territory-planning", "case_count": 6, "personas": [ "Branch Manager", "Regional Sales Director", "Sales Operations Analyst" ], "prompts": [ "Give me a snapshot of my North Valley territory.", "How big is the staffing market in Granite Ridge?", "Where are our biggest coverage gaps?" ] }, "_catalog_kind": "solution", "_readiness": [ "Demo proven", "Conversation tested", "Synthetic data", "Production pattern" ] }, { "schema": "rapp-agent/1.0", "name": "@aibast-agents-library/time-entry-billing", "version": "1.1.0", "display_name": "Time and Entry Billing Agent", "description": "Automate month-end billing cycles to accelerate invoicing, reduce risk, and ensure audit-ready compliance.", "author": "AIBAST", "tags": [ "billing", "time-entry", "invoicing", "audit", "professional-services" ], "category": "professional_services", "quality_tier": "verified", "requires_env": [], "dependencies": [ "@rapp/basic-agent" ], "_file": "agents/@aibast-agents-library/professional_services_stacks/time_entry_billing_stack/time_entry_billing_agent.py", "_size_kb": 36.6, "_lines": 679, "_sha256": "e81e8fb845417641f3c069a673f46a528bba27269ecfc6692038673ec5c0323c", "_install_prefix": "time_entry_billing__", "_install_filename": "time_entry_billing__e81e8fb84541_agent.py", "_stack": "time_entry_billing", "_stack_vertical": "professional_services", "_synthetic_data": true, "_added_at": "2026-03-18T13:23:12-04:00", "_solution": { "advertised_name": "Time Entry and Billing Agent", "sharepoint_list_name": "Month-End Billing Agent", "slot": 8, "executive_summary": "Automate month-end billing cycles to accelerate invoicing, reduce risk, and ensure audit-ready compliance.", "industries": [ "Professional Services, Consulting" ], "personas": [ "Finance VP", "Billing Manager" ], "featured_tools": [ "Dynamics 365", "SharePoint", "Microsoft Teams" ], "agent_requirements": [ "Microsoft Copilot Studio" ], "capabilities": [ "Validate time entries/billing rules and generate invoices", "Complete missing descriptions and classify billable hours", "Flag disputed hours, premium billing, and budget overruns", "Generate CFO-ready revenue recognition reports and audit trails" ], "outcomes": [ "Reduced billing cycles from multi-day process to minutes", "Achieved a higher billable rate via accurate classification", "Protected revenue by recovering disputed hours and preventing leakage", "Delivered audit-ready invoices and compliance documentation" ], "customer_scenario": [ "Month-end billing involves thousands of time entries and manual compliance checks", "Finance leaders face tight deadlines and revenue risk from disputes", "Manual invoice creation slows close cycles and increases error rates" ], "slide_title": "Time Entry and Billing Agent", "slide_summary": "Automate month-end billing cycles to accelerate invoicing, reduce risk, and ensure audit-ready compliance.", "scenario_name": "A professional services firm", "challenge_intro": "A professional services firm faced month-end billing complexity that created risks and delays:", "actions_intro": "The firm used the Time Entry and Billing Agent to:", "outcomes_intro": "With the Agent, the firm:", "opportunity_statements": [ "Process time entries intelligently to enforce billing rules and accelerate invoice generation", "Automate risk and compliance checks to reduce errors and maintain audit-ready documentation", "Provide dispute resolution intelligence to help protect revenue and preserve client relationships" ], "module_name": null, "module_description": null, "source_verified": true, "sales_headline": "Turn month-end billing exceptions into an approval-ready close", "card_pitch": "Help finance and billing teams find blocked work, enforce time-entry rules, prepare support for eligible invoices, and resolve disputed-hour evidence without posting revenue or contacting a client.", "why_try": "The demo shows exactly why work is blocked and separates invoice-ready support from entries that still need narrative, approval, or milestone evidence.", "customer_challenge_copy": "Professional services billing combines large time volumes, contract-specific rules, budget pressure, premium work, and client disputes. Manual review makes it easy to miss evidence, overstate invoice readiness, or treat fixed-fee time like time-and-materials billing.", "microsoft_ai_story": "Microsoft Copilot Studio is the finance-facing entry point. Dynamics 365 provides project and billing context, SharePoint holds approval evidence, and Microsoft Teams supports governed exception review.", "business_value": [ "Improves visibility into billing blockers and leakage.", "Strengthens invoice support and audit consistency.", "Gives finance teams a safer path for resolving disputed hours." ], "search_terms": [ "time entry audit", "month-end billing", "invoice preparation", "unbilled hours", "billing dispute", "fixed fee", "revenue leakage", "Dynamics 365", "SharePoint approvals", "Microsoft Teams", "Copilot Studio" ], "journey_stage": "Operate", "blueprint_role": "Creates an approval-gated billing control point between time capture, invoice support, and client-facing finance work.", "sample_prompts": [ { "label": "Find close blockers", "prompt": "What billable work is still blocked from this close, and what has to happen before it can move?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=billing-blockers" }, { "label": "Review the billing rollup", "prompt": "Give me the month-end billing rollup by project and consultant without treating it as posted revenue.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=billing-rollup" }, { "label": "Audit time cards", "prompt": "Which time cards would fail our billing review because of narrative, hours, rate, or budget concerns?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=billing-audit" }, { "label": "Prepare eligible invoice support", "prompt": "Prepare the invoice support that is actually ready, and keep anything without the right approval or milestone evidence out.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=invoice-support" }, { "label": "Resolve disputed-hour evidence", "prompt": "What evidence is missing on the disputed hours, and what is the safest resolution path before we go back to the clients?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=billing-disputes" } ], "architecture": { "business_flow": [ "Finance and billing teams", "Microsoft Copilot or Copilot Studio", "Time Entry and Billing Agent", "Dynamics 365", "SharePoint", "Microsoft Teams", "Authorized billing review" ], "capabilities": [ { "operation": "unbilled_report", "name": "Billing blocker review", "purpose": "Finds unapproved billable work and the evidence needed to move it." }, { "operation": "billing_summary", "name": "Close rollup", "purpose": "Summarizes project and consultant activity without claiming posted revenue." }, { "operation": "time_entry_audit", "name": "Time-entry control checks", "purpose": "Flags narrative, hours, rate, approval, and budget conditions." }, { "operation": "invoice_preparation", "name": "Approval-gated invoice support", "purpose": "Groups eligible time-and-materials evidence and holds fixed-fee work for milestones." }, { "operation": "dispute_resolution", "name": "Disputed-hours evidence brief", "purpose": "Explains evidence gaps and the internal approval path before client outreach." } ], "easy_mode": [ "Initialize a modern Copilot Studio project.", "Use the deployment recipe and blueprint to author the agent.", "Upload the synthetic ledger and billing-rule knowledge files.", "Bind approved Dynamics 365, SharePoint, and Microsoft Teams connections.", "Validate every locked prompt and stop before publishing." ], "local_install_prompt": "You are GitHub Copilot running in Agent mode in VS Code. Install and open the AI BAST Time Entry and Billing Agent from https://raw.githubusercontent.com/microsoft/aibast-agents-library/main/solutions/time-entry-billing/deployment.json. Verify the source and expected tool, run the smoke prompt, and open the local UI. Do not post revenue, change time, send invoices, connect customer systems, or publish. Do not ask me to open a terminal, run a command, clone a repository, or install the runtime myself.", "copilot_studio_prompt": "You are GitHub Copilot running in Agent mode with the Microsoft Copilot Studio plugin. Promote the AI BAST Time Entry and Billing Agent from its deployment recipe. Preserve all operations and approval gates, upload the packaged knowledge and skills, bind approved Dynamics 365, SharePoint, and Microsoft Teams connections, replay all locked prompts, and stop before publish. Stop before publish.", "required_connections": [ "Dynamics 365", "SharePoint", "Microsoft Teams" ], "manual_commands": [ "pac auth create", "pac copilot init --name \"Time Entry and Billing Agent\" --publisher-prefix --authoring-mode cli-copilot --project-dir \"\" --environment \"\"", "pac connection list --environment \"\"", "pac copilot pull --project-dir \"\"", "pac copilot push --project-dir \"\"", "pac copilot publish --bot \"\" --environment \"\"" ], "acceptance_checks": [ "Every advertised billing workflow is represented.", "Unapproved work remains outside invoice support.", "Fixed-fee time remains separate from invoice value without milestone evidence.", "No output claims posting, recognition, approval, or client contact.", "Publishing remains separately authorized." ], "hard_mode": [ "Create the agent manually.", "Upload both synthetic billing knowledge files.", "Create one skill for each operation.", "Add approved finance and collaboration connections.", "Replay every locked case and confirm the no-posting and no-send boundaries." ] }, "curated_copy": true, "source": "catalog", "origin": null, "onepager": "Time and Entry Billing Agent one-pager.pptx", "demo_video": "#8-Time Entry & Billing Agent.mp4", "has_onepager": true, "has_demo_video": true, "promise_coverage": 80, "is_primary": true, "aliases": [], "package": { "slug": "time-entry-billing", "quest_url": "solutions/time-entry-billing/quest.html", "manual_tutorial_url": "solutions/time-entry-billing/manual-tutorial.html", "export_manifest_url": "solutions/time-entry-billing/export-manifest.json" } }, "_demo": { "slug": "time-entry-billing", "case_count": 9, "personas": [ "Billing Compliance Lead", "Billing Manager", "Client Finance Partner", "Finance Vice President" ], "prompts": [ "What billable work is still blocked from this close, and what has to happen before it can move?", "Give me the month-end billing rollup by project and consultant without treating it as posted revenue.", "Which time cards would fail our billing review because of narrative, hours, rate, or budget concerns?" ] }, "_catalog_kind": "solution", "_readiness": [ "Demo proven", "Conversation tested", "Synthetic data", "Production pattern" ] }, { "schema": "rapp-agent/1.0", "name": "@aibast-agents-library/inventory-visibility", "version": "1.1.0", "display_name": "Inventory Visibility Agent", "description": "Provide synthetic cross-channel inventory visibility and draft replenishment, transfer, and allocation recommendations for authorized review.", "author": "AIBAST", "tags": [ "inventory", "stock-management", "replenishment", "omni-channel", "retail" ], "category": "retail_cpg", "quality_tier": "verified", "requires_env": [], "dependencies": [ "@rapp/basic-agent" ], "_file": "agents/@aibast-agents-library/retail_cpg_stacks/inventory_visibility_stack/inventory_visibility_agent.py", "_size_kb": 41.7, "_lines": 864, "_sha256": "53543d42a294b5bebe87d7e55bfcaeaf4099937398ee761d90b58462ba44aadb", "_install_prefix": "inventory_visibility__", "_install_filename": "inventory_visibility__53543d42a294_agent.py", "_stack": "inventory_visibility", "_stack_vertical": "retail_cpg", "_synthetic_data": true, "_added_at": "2026-03-18T13:23:12-04:00", "_solution": { "advertised_name": "Inventory Visibility Agent", "sharepoint_list_name": "Omnichannel Inventory Agent", "slot": 31, "executive_summary": "Deliver real-time cross-channel inventory intelligence to prevent stockouts, reduce overstock, and maximize omnichannel retail performance.", "industries": [ "Retail" ], "personas": [ "Inventory Planners", "Store Managers", "Category Managers" ], "featured_tools": [ "Dynamics 365 Commerce", "Microsoft Teams" ], "agent_requirements": [ "Microsoft Copilot Studio" ], "capabilities": [ "Identify real\u2011time stock levels across stores, warehouses, and ecommerce", "Detect critical stockouts and instantly locate nearest available inventory", "Recommend same\u2011day transfers and scheduled replenishment", "Evaluate category health to flag overstock, slow movers, and seasonal demand patterns" ], "outcomes": [ "Resolved urgent stockouts quickly and avoided lost sales", "Identified excess inventory and highlighted opportunities", "Enhanced demand planning with more reliable forecasts", "Improved inventory efficiency and accelerated store decision\u2011making" ], "customer_scenario": [ "Store teams spent hours manually checking inventory availability", "Popular products often hit critical stock levels, resulting in lost sales", "Overstocks tied up capital across slow\u2011moving SKUs", "Transfers and replenishment decisions lacked data\u2011driven prioritization" ], "slide_title": "Inventory Visibility Agent", "slide_summary": "Delivers real-time cross-channel inventory intelligence to prevent stockouts, reduce overstock, and maximize omnichannel retail performance.", "scenario_name": "A regional electronics retailer", "challenge_intro": "A regional electronics retailer struggled with stockouts and excess inventory:", "actions_intro": "The retailer used the Inventory Visibility Agent to:", "outcomes_intro": "With the agent\u2019s help, the retailer:", "opportunity_statements": [ "Provide instant SKU\u2011level availability across locations to eliminate manual inventory checks", "Recommend optimal store\u2011to\u2011store and warehouse transfers to ensure products are in the right place at the right time", "Deliver predictive replenishment plans with dynamic safety stock to minimize stockouts and maintain availability" ], "module_name": null, "module_description": null, "source_verified": true, "sales_headline": "Give planners and stores a common inventory picture without making an inventory move.", "card_pitch": "Provides synthetic location and SKU visibility, exception review, replenishment scenarios, and channel planning while keeping every quantity read-only.", "why_try": "It demonstrates how inventory teams can prioritize review with clear assumptions before authorized systems execute any transfer, replenishment, or allocation.", "customer_challenge_copy": "Teams often spend too much time reconciling availability and too little time evaluating stockout, overstock, lead-time, and channel tradeoffs.", "microsoft_ai_story": "The portable agent converts ERP- and commerce-style synthetic records into role-specific planning views suitable for operational review in Teams.", "business_value": [ "Faster exception review", "More transparent replenishment assumptions", "Clearer category and channel tradeoffs" ], "search_terms": [ "inventory visibility", "stock alerts", "replenishment", "channel allocation", "retail planning" ], "journey_stage": "Plan and fulfill", "blueprint_role": "Read-only inventory decision support", "sample_prompts": [ { "label": "Review an inventory snapshot", "prompt": "Summarize the store snapshot and verification boundary.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=inventory-visibility-inventory-dashboard" }, { "label": "Inspect stock exceptions", "prompt": "Identify urgent review candidates without issuing transfer or replenishment commands.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=inventory-visibility-stock-alerts" }, { "label": "Model replenishment", "prompt": "Show the replenishment scenario and its assumptions.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=inventory-visibility-replenishment-plan" }, { "label": "Compare channel allocation", "prompt": "Compare the channel planning scenario without reserving any stock.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=inventory-visibility-channel-allocation" } ], "architecture": { "business_flow": [ "Inventory Planner, Store Manager, and Category Manager", "Microsoft 365 Copilot or Copilot Studio", "Inventory Visibility Agent", "Approved read-only ERP, commerce, and warehouse data", "Microsoft Teams review", "Authorized inventory systems" ], "capabilities": [ { "operation": "inventory_dashboard", "name": "Read-only inventory visibility", "purpose": "Shows location and SKU evidence with system-of-record verification." }, { "operation": "stock_alerts", "name": "Stock review candidates", "purpose": "Identifies exceptions without issuing transfer or replenishment commands." }, { "operation": "replenishment_plan", "name": "Draft replenishment scenario", "purpose": "Models quantities, sources, lead times, and cost without creating orders." }, { "operation": "channel_allocation", "name": "Channel allocation scenario", "purpose": "Compares planning weights without reserving or allocating stock." } ], "easy_mode": [ "Initialize a modern Copilot Studio project in the target environment.", "Provide this blueprint and the deployment recipe to the Copilot Studio Architect workflow.", "Upload the packaged global instructions, synthetic knowledge files, and operation skills.", "Bind only approved read-only connections and keep all write actions separate.", "Push the draft, replay every locked prompt, and stop before publishing." ], "local_install_prompt": "You are GitHub Copilot running in Agent mode in VS Code. Install and open the AI BAST Inventory Visibility Agent from https://raw.githubusercontent.com/microsoft/aibast-agents-library/main/solutions/inventory-visibility/deployment.json. Own the setup, verify the source path, confirm the expected tool in Brainstem health, run the smoke prompt, and open the local UI. Use synthetic data only, preserve every safety boundary, and do not publish or silently substitute another agent. Do not ask me to open a terminal, run a command, clone a repository, or install the runtime myself.", "copilot_studio_prompt": "You are GitHub Copilot running in Agent mode with the Microsoft Copilot Studio plugin. Prepare a draft Inventory Visibility Agent from its deployment recipe. Ask only for the target environment and publisher prefix. Upload the packaged global instructions, two knowledge files, and one skill per operation; bind approved read-only connections; validate all locked prompts; and stop before publish with every external action and authorization gate visible. Stop before publish.", "required_connections": [ "Dynamics Commerce or ERP through an approved read-only integration", "Microsoft Teams for planning review" ], "manual_commands": [ "pac auth create", "pac copilot init --name \"Inventory Visibility Agent\" --publisher-prefix --authoring-mode cli-copilot --project-dir \"\" --environment \"\"", "pac connection list --environment \"\"", "pac copilot pull --project-dir \"\"", "pac copilot push --project-dir \"\"", "pac copilot publish --bot \"\" --environment \"\"" ], "acceptance_checks": [ "Every advertised capability is represented by a deterministic operation and uploadable skill.", "The exact portable source stem is present in the capture contract.", "Both synthetic knowledge files and the global instructions are packaged under the standard solution directory.", "Outputs preserve privacy, recommendation-only behavior, and explicit human approval gates.", "All locked persona prompts route to deterministic synthetic evidence.", "Publishing remains a separate authorized step." ], "hard_mode": [ "Create the agent manually in Copilot Studio.", "Enter the packaged global instructions.", "Upload both synthetic knowledge files.", "Create one skill for each operation.", "Bind only approved read-only connections.", "Replay every locked case and verify safety gates before publishing." ] }, "curated_copy": true, "source": "catalog", "origin": null, "onepager": "34. Inventory Visibility Agent one-pager.pptx", "demo_video": "34-Inventory Visibility Agent.mp4", "has_onepager": true, "has_demo_video": true, "promise_coverage": 65, "is_primary": true, "aliases": [], "package": { "slug": "inventory-visibility", "quest_url": "solutions/inventory-visibility/quest.html", "manual_tutorial_url": "solutions/inventory-visibility/manual-tutorial.html", "export_manifest_url": "solutions/inventory-visibility/export-manifest.json" } }, "_demo": { "slug": "inventory-visibility", "case_count": 8, "personas": [ "Category Manager", "Inventory Planner", "Store Manager" ], "prompts": [ "As Inventory Planner, summarize the store snapshot and verification boundary.", "As Store Manager, identify urgent review candidates without issuing transfer or replenishment commands.", "As Inventory Planner, show the fourteen-day replenishment scenario and its assumptions." ] }, "_catalog_kind": "solution", "_readiness": [ "Demo proven", "Conversation tested", "Synthetic data", "Production pattern" ] }, { "schema": "rapp-agent/1.0", "name": "@aibast-agents-library/personalized-marketing", "version": "1.0.0", "display_name": "Personalized Marketing Agent", "description": "Draft privacy-safe customer segment insights, multi-wave campaign plans, personalized content with A/B variants, automation workflows, revenue projections, and executive briefs for human review.", "author": "AIBAST", "tags": [ "marketing", "personalization", "segmentation", "campaigns", "retail" ], "category": "retail_cpg", "quality_tier": "verified", "requires_env": [], "dependencies": [ "@rapp/basic-agent" ], "_file": "agents/@aibast-agents-library/retail_cpg_stacks/personalized_marketing_stack/personalized_marketing_agent.py", "_size_kb": 29.8, "_lines": 706, "_sha256": "c4d163da8e0e85cf4e84398a16acf5cf16f0fb6d13d157c7c2460f46d1e49eb9", "_install_prefix": "personalized_marketing__", "_install_filename": "personalized_marketing__c4d163da8e0e_agent.py", "_stack": "personalized_marketing", "_stack_vertical": "retail_cpg", "_synthetic_data": true, "_added_at": "2026-03-18T13:23:12-04:00", "_solution": { "advertised_name": "Personalized Marketing Agent", "sharepoint_list_name": "Campaign Design Agent", "slot": 14, "executive_summary": "Automate personalized campaign design and execution to boost engagement, accelerate revenue, and strengthen customer loyalty.", "industries": [ "Retail" ], "personas": [ "Marketing Director", "Campaign Manager" ], "featured_tools": [ "Dynamics 365", "Microsoft Teams" ], "agent_requirements": [ "Microsoft Copilot Studio" ], "capabilities": [ "Automatically segment customers based on real behavior and value signals", "Design multi-wave campaigns with tailored messaging, offers, and timing", "Continuously optimize engagement through automated testing and workflows" ], "outcomes": [ "Drove higher conversion rates via precision-targeted campaigns", "Reduced campaign planning cycles from weeks to hours", "Increased repeat purchases and average order value", "Provided leaders with real-time performance insights to guide decisions during peak periods" ], "customer_scenario": [ "Manual campaign planning took weeks, slowing response to market signals", "Messaging was underperforming with high-value customer segments", "Marketers lacked real-time insight into campaign performance and impact" ], "slide_title": "Personalized Marketing Agent", "slide_summary": "Automate personalized campaign design and execution to boost engagement, accelerate revenue, and strengthen customer loyalty.", "scenario_name": "A large retail company", "challenge_intro": "A retail company faced a high holiday revenue target and declining customer engagement.", "actions_intro": "The retail company used the Personalized Marketing Agent to:", "outcomes_intro": "With the agent\u2019s help, the retailer:", "opportunity_statements": [ "Perform advanced customer segmentation to target high-value groups and maximize ROI", "Generate multi-wave personalized campaigns to drive higher conversion rates and accelerate revenue growth", "Deliver A/B testing and scheduling workflow orchestration to streamline execution and marketing team efficiency" ], "module_name": null, "module_description": null, "source_verified": true, "sales_headline": "Move from slow campaign planning to governed, review-ready audience and content decisions.", "card_pitch": "Helps retail marketing teams understand aggregate commercial segments, compare campaign concepts, draft relevant content, and review synthetic performance evidence without sensitive profiling or activation.", "why_try": "It gives leaders and campaign operators a common planning surface while keeping audience selection, copy, incentives, and launch decisions behind human approval.", "customer_challenge_copy": "Retail marketing teams often work across fragmented aggregate signals, long planning cycles, inconsistent content, and delayed performance insight.", "microsoft_ai_story": "A portable agent turns synthetic CRM-style segment and campaign records into governed planning briefs that can be reviewed in Microsoft collaboration workflows before any production activation.", "business_value": [ "Shorter planning discussions with clearer assumptions", "More consistent review of audience, content, and measurement choices", "Stronger privacy and approval boundaries for personalization" ], "search_terms": [ "retail marketing", "campaign planning", "customer segments", "content drafts", "marketing measurement" ], "journey_stage": "Plan and evaluate", "blueprint_role": "Privacy-safe marketing decision support", "sample_prompts": [ { "label": "Review aggregate segments", "prompt": "Summarize the aggregate customer groups without demographic traits and identify portfolio review priorities.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=personalized-marketing-customer-segmentation" }, { "label": "Design a governed campaign concept", "prompt": "Outline the review-only win-back sequence, assumptions, and approval gates.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=personalized-marketing-campaign-design" }, { "label": "Draft privacy-safe content", "prompt": "Draft neutral content ideas for New Explorers and explain the non-sensitive signals used.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=personalized-marketing-content-personalization" }, { "label": "Evaluate synthetic performance", "prompt": "Compare the synthetic tests and call out measurement limitations before any decision.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=personalized-marketing-performance-analysis" } ], "architecture": { "business_flow": [ "Marketing Director and Campaign Manager", "Microsoft 365 Copilot or Copilot Studio", "Personalized Marketing Agent", "Approved read-only CRM or customer-insights data", "Microsoft Teams review", "Authorized marketing approval" ], "capabilities": [ { "operation": "customer_segmentation", "name": "Aggregate customer segmentation", "purpose": "Compares non-sensitive commercial segment signals without identifying or profiling a person." }, { "operation": "campaign_design", "name": "Campaign concept design", "purpose": "Builds review-only multi-wave campaign concepts with approval gates." }, { "operation": "content_personalization", "name": "Draft content personalization", "purpose": "Produces segment-level draft copy and product ideas from non-sensitive aggregate signals." }, { "operation": "performance_analysis", "name": "Synthetic performance analysis", "purpose": "Compares fixed test and campaign scenarios without claiming live results." } ], "easy_mode": [ "Initialize a modern Copilot Studio project in the target environment.", "Provide this blueprint and the deployment recipe to the Copilot Studio Architect workflow.", "Upload the packaged global instructions, synthetic knowledge files, and operation skills.", "Bind only approved read-only connections and keep all write actions separate.", "Push the draft, replay every locked prompt, and stop before publishing." ], "local_install_prompt": "You are GitHub Copilot running in Agent mode in VS Code. Install and open the AI BAST Personalized Marketing Agent from https://raw.githubusercontent.com/microsoft/aibast-agents-library/main/solutions/personalized-marketing/deployment.json. Own the setup, verify the source path, confirm the expected tool in Brainstem health, run the smoke prompt, and open the local UI. Use synthetic data only, preserve every safety boundary, and do not publish or silently substitute another agent. Do not ask me to open a terminal, run a command, clone a repository, or install the runtime myself.", "copilot_studio_prompt": "You are GitHub Copilot running in Agent mode with the Microsoft Copilot Studio plugin. Prepare a draft Personalized Marketing Agent from its deployment recipe. Ask only for the target environment and publisher prefix. Upload the packaged global instructions, two knowledge files, and one skill per operation; bind approved read-only connections; validate all locked prompts; and stop before publish with every external action and authorization gate visible. Stop before publish.", "required_connections": [ "Dynamics CRM data through an approved read-only integration", "Microsoft Teams for human review" ], "manual_commands": [ "pac auth create", "pac copilot init --name \"Personalized Marketing Agent\" --publisher-prefix --authoring-mode cli-copilot --project-dir \"\" --environment \"\"", "pac connection list --environment \"\"", "pac copilot pull --project-dir \"\"", "pac copilot push --project-dir \"\"", "pac copilot publish --bot \"\" --environment \"\"" ], "acceptance_checks": [ "Every advertised capability is represented by a deterministic operation and uploadable skill.", "The exact portable source stem is present in the capture contract.", "Both synthetic knowledge files and the global instructions are packaged under the standard solution directory.", "Outputs preserve privacy, recommendation-only behavior, and explicit human approval gates.", "All locked persona prompts route to deterministic synthetic evidence.", "Publishing remains a separate authorized step." ], "hard_mode": [ "Create the agent manually in Copilot Studio.", "Enter the packaged global instructions.", "Upload both synthetic knowledge files.", "Create one skill for each operation.", "Bind only approved read-only connections.", "Replay every locked case and verify safety gates before publishing." ] }, "curated_copy": true, "source": "catalog", "origin": null, "onepager": "Personalized Marketing Agent one-pager.pptx", "demo_video": "19-Personalized Marketing Agent.mp4", "has_onepager": true, "has_demo_video": true, "promise_coverage": 75, "is_primary": true, "aliases": [], "package": { "slug": "personalized-marketing", "quest_url": "solutions/personalized-marketing/quest.html", "manual_tutorial_url": "solutions/personalized-marketing/manual-tutorial.html", "export_manifest_url": "solutions/personalized-marketing/export-manifest.json" } }, "_demo": { "slug": "personalized-marketing", "case_count": 7, "personas": [ "Campaign Manager", "Marketing Director" ], "prompts": [ "As Marketing Director, summarize the aggregate customer groups without demographic traits and identify portfolio review priorities.", "As Campaign Manager, show me the personalized multi-wave campaign recommendations for the holiday promotion and the approval gates.", "As Campaign Manager, generate the VIP campaign with personalized content and A/B test variants." ] }, "_catalog_kind": "solution", "_readiness": [ "Demo proven", "Conversation tested", "Synthetic data", "Production pattern" ] }, { "schema": "rapp-agent/1.0", "name": "@aibast-agents-library/returns-complaints-resolution", "version": "1.0.0", "display_name": "Returns and Complaints Resolution Agent", "description": "Draft privacy-safe return reviews, complaint classifications, resolution options, and aggregate trend analysis for authorized human decision-makers.", "author": "AIBAST", "tags": [ "returns", "complaints", "customer-service", "resolution", "retail" ], "category": "retail_cpg", "quality_tier": "verified", "requires_env": [], "dependencies": [ "@rapp/basic-agent" ], "_file": "agents/@aibast-agents-library/retail_cpg_stacks/returns_complaints_resolution_stack/returns_complaints_resolution_agent.py", "_size_kb": 40.9, "_lines": 904, "_sha256": "62811c3bb8c178d32230efabdd85ec151e2fee400e4bfd8bff75a35d3a414fa3", "_install_prefix": "returns_complaints_resolution__", "_install_filename": "returns_complaints_resolution__62811c3bb8c1_agent.py", "_stack": "returns_complaints_resolution", "_stack_vertical": "retail_cpg", "_synthetic_data": true, "_added_at": "2026-03-18T13:23:12-04:00", "_solution": { "advertised_name": "Returns and Complaints Resolution Agent", "sharepoint_list_name": "Returns & Complaints Agent", "slot": 41, "executive_summary": "Automate return decisions and complaint handling to speed resolution, reduce fraud, and protect customer loyalty.", "industries": [ "Retail", "Consumable Products Industry" ], "personas": [ "Customer Service Agents", "Quality Teams", "Loss Prevention Teams" ], "featured_tools": [ "Dynamics 365, Outlook,", "SharePoint, Teams" ], "agent_requirements": [ "Microsoft Copilot Studio" ], "capabilities": [ "Automatically evaluate and classify return requests", "Prioritize complaints with high sentiment, draft tailored responses, and execute resolution options", "Analyze SKU patterns for quality issues and suspicious behavior", "Recommend resolution strategies based on customer value" ], "outcomes": [ "Accelerated return and complaint resolution", "Delivered consistent, empathetic customer communications", "Reduced fraudulent returns and improved early defect detection", "Increased customer retention by applying recovery strategies that preserve lifetime value" ], "customer_scenario": [ "Manual review steps led to inconsistent refund and replacement decisions", "Frustrated customers escalated complaints due to slow responses and unclear next steps", "Product defects and fraudulent return patterns were identified too late to prevent losses" ], "slide_title": "Returns and Complaints Resolution Agent", "slide_summary": "Automates return decisions and complaint handling to speed resolution, reduce fraud, and protect customer loyalty.", "scenario_name": "An eCommerce company", "challenge_intro": "An eCommerce company faced rising return volumes and customer complaints across channels.", "actions_intro": "The company used the Returns and Complaints Resolution Agent to:", "outcomes_intro": "With the Agent, the company:", "opportunity_statements": [ "Evaluate returns to improve decision accuracy and resolution efficiency", "Analyze sentiment and draft personalized responses to strengthen recovery outcomes", "Detect product patterns to identify quality issues and reduce fraud", "Execute resolution actions and recommend strategies to streamline service workflows" ], "module_name": null, "module_description": null, "source_verified": true, "sales_headline": "Bring consistency and empathy to return review without automating the decision.", "card_pitch": "Summarizes anonymous synthetic return evidence, classifies complaint themes, drafts policy-grounded options, and surfaces aggregate quality trends.", "why_try": "It helps service, quality, and loss-prevention teams review the same evidence while preserving authorization, privacy, and non-accusatory language.", "customer_challenge_copy": "Return and complaint workflows can become inconsistent, slow, and difficult to learn from when policy evidence, product signals, and service context are fragmented.", "microsoft_ai_story": "The agent converts CRM- and ERP-style synthetic case records into structured decision support for human review in Microsoft collaboration workflows.", "business_value": [ "More consistent review summaries", "Faster visibility into aggregate quality patterns", "Clearer approval and communication boundaries" ], "search_terms": [ "returns", "complaints", "customer service", "quality trends", "resolution options" ], "journey_stage": "Resolve and learn", "blueprint_role": "Human-approved service decision support", "sample_prompts": [ { "label": "Review an anonymous return case", "prompt": "Summarize the anonymous return review evidence and approval boundary.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=returns-complaints-resolution-return-processing" }, { "label": "Classify a complaint safely", "prompt": "Classify this product concern without echoing personal information or sending a response.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=returns-complaints-resolution-complaint-classification" }, { "label": "Draft a resolution option", "prompt": "Draft a policy-grounded option and keep all actions behind authorization.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=returns-complaints-resolution-resolution-recommendation" }, { "label": "Analyze aggregate quality trends", "prompt": "Summarize aggregate defect and return patterns without accusing any person.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=returns-complaints-resolution-trend-analysis" } ], "architecture": { "business_flow": [ "Customer Service Agent, Quality Team, and Loss Prevention Team", "Microsoft 365 Copilot or Copilot Studio", "Returns and Complaints Resolution Agent", "Approved read-only CRM, order, and policy evidence", "SharePoint governed policy knowledge", "Authorized service resolution workflow" ], "capabilities": [ { "operation": "return_processing", "name": "Anonymous return review", "purpose": "Summarizes policy evidence without approving or processing a return." }, { "operation": "complaint_classification", "name": "Privacy-safe complaint classification", "purpose": "Classifies a concern without echoing personal or sensitive text." }, { "operation": "resolution_recommendation", "name": "Human-approved resolution options", "purpose": "Drafts policy-grounded options without refund, credit, replacement, shipment, or messaging actions." }, { "operation": "trend_analysis", "name": "Aggregate returns and quality trends", "purpose": "Surfaces product and service patterns without accusing an individual." } ], "easy_mode": [ "Initialize a modern Copilot Studio project in the target environment.", "Provide this blueprint and the deployment recipe to the Copilot Studio Architect workflow.", "Upload the packaged global instructions, synthetic knowledge files, and operation skills.", "Bind only approved read-only connections and keep all write actions separate.", "Push the draft, replay every locked prompt, and stop before publishing." ], "local_install_prompt": "You are GitHub Copilot running in Agent mode in VS Code. Install and open the AI BAST Returns and Complaints Resolution Agent from https://raw.githubusercontent.com/microsoft/aibast-agents-library/main/solutions/returns-complaints-resolution/deployment.json. Own the setup, verify the source path, confirm the expected tool in Brainstem health, run the smoke prompt, and open the local UI. Use synthetic data only, preserve every safety boundary, and do not publish or silently substitute another agent. Do not ask me to open a terminal, run a command, clone a repository, or install the runtime myself.", "copilot_studio_prompt": "You are GitHub Copilot running in Agent mode with the Microsoft Copilot Studio plugin. Prepare a draft Returns and Complaints Resolution Agent from its deployment recipe. Ask only for the target environment and publisher prefix. Upload the packaged global instructions, two knowledge files, and one skill per operation; bind approved read-only connections; validate all locked prompts; and stop before publish with every external action and authorization gate visible. Stop before publish.", "required_connections": [ "Dynamics CRM and ERP through approved read-only integrations", "SharePoint for governed policy knowledge" ], "manual_commands": [ "pac auth create", "pac copilot init --name \"Returns and Complaints Resolution Agent\" --publisher-prefix --authoring-mode cli-copilot --project-dir \"\" --environment \"\"", "pac connection list --environment \"\"", "pac copilot pull --project-dir \"\"", "pac copilot push --project-dir \"\"", "pac copilot publish --bot \"\" --environment \"\"" ], "acceptance_checks": [ "Every advertised capability is represented by a deterministic operation and uploadable skill.", "The exact portable source stem is present in the capture contract.", "Both synthetic knowledge files and the global instructions are packaged under the standard solution directory.", "Outputs preserve privacy, recommendation-only behavior, and explicit human approval gates.", "All locked persona prompts route to deterministic synthetic evidence.", "Publishing remains a separate authorized step." ], "hard_mode": [ "Create the agent manually in Copilot Studio.", "Enter the packaged global instructions.", "Upload both synthetic knowledge files.", "Create one skill for each operation.", "Bind only approved read-only connections.", "Replay every locked case and verify safety gates before publishing." ] }, "curated_copy": true, "source": "catalog", "origin": null, "onepager": "51. Returns and Complaints Resolution Agent one-pager 2026 02 23.pptx", "demo_video": "51-Returns and Complaints Resolution Agent.mp4", "has_onepager": true, "has_demo_video": true, "promise_coverage": 45, "is_primary": true, "aliases": [], "package": { "slug": "returns-complaints-resolution", "quest_url": "solutions/returns-complaints-resolution/quest.html", "manual_tutorial_url": "solutions/returns-complaints-resolution/manual-tutorial.html", "export_manifest_url": "solutions/returns-complaints-resolution/export-manifest.json" } }, "_demo": { "slug": "returns-complaints-resolution", "case_count": 10, "personas": [ "Customer Service Agent", "Quality Team", "Service Manager" ], "prompts": [ "As Customer Service Agent, summarize the anonymous return review evidence and approval boundary.", "As Customer Service Agent, classify this product concern without echoing personal information or sending a response.", "As Customer Service Agent, draft a policy-grounded option and keep all actions behind authorization." ] }, "_catalog_kind": "solution", "_readiness": [ "Demo proven", "Conversation tested", "Synthetic data", "Production pattern" ] }, { "schema": "rapp-agent/1.0", "name": "@aibast-agents-library/store-associate-copilot", "version": "1.0.0", "display_name": "Retail Store Associate Copilot", "description": "Provide synthetic product intelligence, customer-assistance drafts, task planning, and aggregate coaching insights for associate review.", "author": "AIBAST", "tags": [ "store-operations", "associate", "copilot", "product-lookup", "retail" ], "category": "retail_cpg", "quality_tier": "verified", "requires_env": [], "dependencies": [ "@rapp/basic-agent" ], "_file": "agents/@aibast-agents-library/retail_cpg_stacks/store_associate_copilot_stack/store_associate_copilot_agent.py", "_size_kb": 35.4, "_lines": 794, "_sha256": "657abd8d24792ddf5b9e8abcad7eccf0ad5c0f3fa5f3ccb94c8440532a50ae31", "_install_prefix": "store_associate_copilot__", "_install_filename": "store_associate_copilot__657abd8d2479_agent.py", "_stack": "store_associate_copilot", "_stack_vertical": "retail_cpg", "_synthetic_data": true, "_added_at": "2026-03-18T13:23:12-04:00", "_solution": { "advertised_name": "Retail Store Associate Copilot", "sharepoint_list_name": "Store Associate Agent", "slot": 21, "executive_summary": "Deliver real-time product intelligence and transaction support to deliver faster service and boost sales performance.", "industries": [ "Retail" ], "personas": [ "Store Associate", "Sales Manager", "Floor Specialist" ], "featured_tools": [ "Dynamics 365", "Microsoft Teams" ], "agent_requirements": [ "Microsoft Copilot Studio" ], "capabilities": [ "Surface product specs, comparisons, and current promotions instantly", "Confirm inventory and suggest alternatives for out-of-stock items", "Recommend accessories and warranties with conversion tips", "Prepare transactions with loyalty discounts and financing options automatically" ], "outcomes": [ "Reduced interaction time dramatically", "Increased accessory attach rates and average transaction value", "Eliminated pricing errors and improved checkout speed", "Enabled associates to serve more customers and boost commission" ], "customer_scenario": [ "Associates spent minutes searching for product details and stock levels", "Manual promotion application caused pricing errors and delays", "Upsell opportunities were missed due to lack of real-time guidance", "Customer satisfaction scores declined during peak traffic periods" ], "slide_title": "Retail Store Associate Copilot", "slide_summary": "Delivers real-time product intelligence and transaction support to deliver faster service and boost sales performance.", "scenario_name": "A consumer electronics retailer", "challenge_intro": "A consumer electronics retailer needed to address long customer wait times and low sales numbers:", "actions_intro": "The retailer leveraged the Retail Store Associate Copilot to:", "outcomes_intro": "With the agent\u2019s help, the retailer:", "opportunity_statements": [ "Provide instant access to product details to accelerate customer interactions", "Display real-time inventory and availability to enhance customer trust", "Recommend add-ons and upsell opportunities to increase transaction value", "Automate transaction preparation to reduce friction and improve conversion" ], "module_name": null, "module_description": null, "source_verified": true, "sales_headline": "Put consistent product guidance and floor planning at the associate\u2019s fingertips.", "card_pitch": "Provides synthetic product facts, respectful customer-language drafts, shift planning, and role-cohort coaching signals while leaving availability, pricing, returns, and transactions to authorized systems and people.", "why_try": "It demonstrates how associates can get faster context without creating inventory promises, pressure-based selling, or workforce profiling.", "customer_challenge_copy": "Store teams lose time searching for product details, coordinating floor work, and translating service expectations into consistent customer conversations.", "microsoft_ai_story": "The portable agent combines catalog-style records and store planning data into concise associate and manager views suitable for collaborative review.", "business_value": [ "Faster access to consistent product context", "Clearer shift planning and service language", "Safer aggregate coaching without personnel decisions" ], "search_terms": [ "store associate", "product lookup", "floor operations", "customer assistance", "retail coaching" ], "journey_stage": "Serve and coordinate", "blueprint_role": "Associate guidance with transaction guardrails", "sample_prompts": [ { "label": "Look up a product safely", "prompt": "Give me the product facts, floor location, optional complements, and availability caveat for the running shoes.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=store-associate-copilot-product-lookup" }, { "label": "Draft customer-assistance language", "prompt": "Draft respectful language for acknowledging a complaint while keeping the decision with an authorized reviewer.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=store-associate-copilot-customer-assist" }, { "label": "Plan the store shift", "prompt": "Turn the opening work into a prioritized planning checklist without claiming execution.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=store-associate-copilot-task-checklist" }, { "label": "Review aggregate coaching signals", "prompt": "Summarize aggregate role-cohort coaching signals without ranking employees.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=store-associate-copilot-performance-dashboard" } ], "architecture": { "business_flow": [ "Store Associate, Floor Specialist, and Sales Manager", "Microsoft 365 Copilot or Copilot Studio", "Retail Store Associate Copilot", "Approved commerce catalog and inventory data", "Microsoft Teams review", "Authorized store and transaction systems" ], "capabilities": [ { "operation": "product_lookup", "name": "Verified product lookup", "purpose": "Surfaces product facts, floor location, and a synthetic availability snapshot with a verification gate." }, { "operation": "customer_assist", "name": "Respectful assistance drafts", "purpose": "Drafts pressure-free customer language without messaging or transacting." }, { "operation": "task_checklist", "name": "Shift planning checklist", "purpose": "Organizes opening, midday, and closing work without claiming execution." }, { "operation": "performance_dashboard", "name": "Role-cohort coaching review", "purpose": "Summarizes aggregate workflow signals without profiling employees or making personnel decisions." } ], "easy_mode": [ "Initialize a modern Copilot Studio project in the target environment.", "Provide this blueprint and the deployment recipe to the Copilot Studio Architect workflow.", "Upload the packaged global instructions, synthetic knowledge files, and operation skills.", "Bind only approved read-only connections and keep all write actions separate.", "Push the draft, replay every locked prompt, and stop before publishing." ], "local_install_prompt": "You are GitHub Copilot running in Agent mode in VS Code. Install and open the AI BAST Retail Store Associate Copilot from https://raw.githubusercontent.com/microsoft/aibast-agents-library/main/solutions/store-associate-copilot/deployment.json. Own the setup, verify the source path, confirm the expected tool in Brainstem health, run the smoke prompt, and open the local UI. Use synthetic data only, preserve every safety boundary, and do not publish or silently substitute another agent. Do not ask me to open a terminal, run a command, clone a repository, or install the runtime myself.", "copilot_studio_prompt": "You are GitHub Copilot running in Agent mode with the Microsoft Copilot Studio plugin. Prepare a draft Retail Store Associate Copilot from its deployment recipe. Ask only for the target environment and publisher prefix. Upload the packaged global instructions, two knowledge files, and one skill per operation; bind approved read-only connections; validate all locked prompts; and stop before publish with every external action and authorization gate visible. Stop before publish.", "required_connections": [ "Dynamics Commerce or ERP through approved read-only integrations", "Microsoft Teams for manager review" ], "manual_commands": [ "pac auth create", "pac copilot init --name \"Retail Store Associate Copilot\" --publisher-prefix --authoring-mode cli-copilot --project-dir \"\" --environment \"\"", "pac connection list --environment \"\"", "pac copilot pull --project-dir \"\"", "pac copilot push --project-dir \"\"", "pac copilot publish --bot \"\" --environment \"\"" ], "acceptance_checks": [ "Every advertised capability is represented by a deterministic operation and uploadable skill.", "The exact portable source stem is present in the capture contract.", "Both synthetic knowledge files and the global instructions are packaged under the standard solution directory.", "Outputs preserve privacy, recommendation-only behavior, and explicit human approval gates.", "All locked persona prompts route to deterministic synthetic evidence.", "Publishing remains a separate authorized step." ], "hard_mode": [ "Create the agent manually in Copilot Studio.", "Enter the packaged global instructions.", "Upload both synthetic knowledge files.", "Create one skill for each operation.", "Bind only approved read-only connections.", "Replay every locked case and verify safety gates before publishing." ] }, "curated_copy": true, "source": "catalog", "origin": null, "onepager": "24. Retail Store Associate Copilot one-pager.pptx", "demo_video": "24-Store Associate Copilot.mp4", "has_onepager": true, "has_demo_video": true, "promise_coverage": 45, "is_primary": true, "aliases": [], "package": { "slug": "store-associate-copilot", "quest_url": "solutions/store-associate-copilot/quest.html", "manual_tutorial_url": "solutions/store-associate-copilot/manual-tutorial.html", "export_manifest_url": "solutions/store-associate-copilot/export-manifest.json" } }, "_demo": { "slug": "store-associate-copilot", "case_count": 7, "personas": [ "Floor Specialist", "Sales Manager", "Store Associate" ], "prompts": [ "As Store Associate, give me the product facts, floor location, optional complements, and availability caveat for SKU-1005.", "As Store Associate, draft respectful language for acknowledging a complaint while keeping the decision with an authorized reviewer.", "As Floor Specialist, turn the opening work into a prioritized planning checklist without claiming execution." ] }, "_catalog_kind": "solution", "_readiness": [ "Demo proven", "Conversation tested", "Synthetic data", "Production pattern" ] }, { "schema": "rapp-agent/1.0", "name": "@aibast-agents-library/supplier-claim-engagement", "version": "1.0.0", "display_name": "Supplier Claim Engagement Agent", "description": "Help retail claims teams recover more from suppliers by triaging damage, shortage, quality, and pricing claims, assembling evidence packs, drafting supplier claim notices, tracking response deadlines, and evaluating supplier responses, while every send, credit acceptance, and escalation stays with an authorized coordinator.", "author": "AIBAST", "tags": [ "retail", "supplier", "claims", "returns", "service-levels", "retail-cpg" ], "category": "retail_cpg", "quality_tier": "verified", "requires_env": [], "dependencies": [ "@rapp/basic-agent" ], "_file": "agents/@aibast-agents-library/retail_cpg_stacks/supplier_claim_engagement_stack/supplier_claim_engagement_agent.py", "_size_kb": 22.4, "_lines": 428, "_sha256": "f4be20770c2bc9e239e911d8e0d914ed031e6b5330c6577c2ed4c9da02a8adf5", "_install_prefix": "supplier_claim_engagement__", "_install_filename": "supplier_claim_engagement__f4be20770c2b_agent.py", "_stack": "supplier_claim_engagement", "_stack_vertical": "retail_cpg", "_synthetic_data": true, "_added_at": "2026-10-08T10:30:56-04:00", "_solution": { "advertised_name": "Supplier Claim Engagement Agent", "sharepoint_list_name": "Supplier Claim Engagement Agent", "slot": 62, "executive_summary": "Help retail claims teams recover more from suppliers by triaging damage, shortage, quality, and pricing claims, assembling evidence packs, drafting supplier claim notices, tracking response deadlines, and evaluating supplier responses, while every send, credit acceptance, and escalation stays with an authorized coordinator.", "industries": [ "Retail", "Consumer Goods" ], "personas": [ "Supplier Claims Coordinator", "Receiving Supervisor", "Claims Team Lead", "Category Manager" ], "featured_tools": [ "Microsoft Teams", "Microsoft Outlook", "Dynamics 365" ], "agent_requirements": [ "Microsoft Copilot Studio" ], "capabilities": [ "Triages damage, shortage, quality and pricing claims", "Assembles evidence packs and drafts claim notices", "Tracks response windows and drafts escalations before value is lost", "Evaluates supplier responses to surface unrecovered value" ], "outcomes": [ "More value recovered: every open claim and its value in one queue", "Faster filing: evidence packs arrive as ready drafts", "Fewer missed deadlines: at-risk claims surface early" ], "customer_scenario": [ "Damaged, short and off-quality deliveries are logged apart", "Evidence packs are rebuilt by hand from photos and notes", "Supplier response deadlines slip because nobody tracks the clock", "Partial credits are accepted without checking the gap against the claim" ], "slide_title": "Supplier Claim Engagement Agent", "slide_summary": "Helps retail claims teams recover more from suppliers with triage, evidence packs and deadline tracking, while coordinators approve every send.", "scenario_name": "A regional grocery retailer", "challenge_intro": "A grocery retailer's claims team chases suppliers across email and spreadsheets:", "actions_intro": "The Supplier Claim Engagement Agent covers triage through supplier response:", "outcomes_intro": "Claims teams recover more value with less manual chasing:", "opportunity_statements": [ "See every open supplier claim and its value", "File claims with complete evidence the same day", "Catch missed supplier deadlines before value is lost" ], "module_name": null, "module_description": null, "source_verified": true, "sales_headline": "Recover more from suppliers without losing control of what gets sent", "card_pitch": "Give supplier claims teams one conversational view to triage claims, assemble evidence, draft supplier notices, track response deadlines, and weigh supplier offers, while every send, credit decision, and escalation stays with an authorized coordinator.", "why_try": "The demo lets a supplier claims coordinator ask a natural question and receive a source-backed, reviewable answer with the approval boundary visible in the response.", "customer_challenge_copy": "Retail claims teams rebuild evidence by hand, lose track of supplier response deadlines, and accept partial credits without checking the gap.", "microsoft_ai_story": "Copilot Studio and Microsoft Teams give coordinators a conversational front door; approved order, receiving, and case systems can be bound through least-privilege connections, with Outlook sends kept behind human approval.", "business_value": [ "Makes open claim value and recovery visible in one view.", "Turns evidence gathering and supplier notices into reviewable drafts.", "Flags breached and at-risk supplier deadlines with escalation drafts." ], "search_terms": [ "Category Manager", "Claims Team Lead", "Dynamics 365 Customer Service", "Microsoft Copilot Studio", "Microsoft Outlook", "Microsoft Teams", "Receiving Supervisor", "Supplier Claim Engagement Agent", "Supplier Claims Coordinator", "claims_queue", "damaged goods", "draft_supplier_claim", "escalation_brief", "evidence pack", "evidence_pack", "response deadline", "shortage claims", "sla_tracker", "supplier claims", "supplier_response", "triage_claim" ], "journey_stage": "Optimize", "blueprint_role": "Separates claim triage, evidence assembly, and drafting from protected supplier communication and credit decisions.", "sample_prompts": [ { "label": "Supplier Claims Coordinator: Supplier claims queue", "prompt": "Show me our open supplier claims and what needs attention today.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=supplier-claim-engagement-claims-queue" }, { "label": "Supplier Claims Coordinator: Claim triage", "prompt": "Triage the new damage claim CLM-5201. Who is responsible and what are the terms?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=supplier-claim-engagement-triage-claim" }, { "label": "Receiving Supervisor: Evidence pack check", "prompt": "Is the evidence for CLM-5201 complete, and what is the claim worth?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=supplier-claim-engagement-evidence-pack" }, { "label": "Supplier Claims Coordinator: Supplier claim notice draft", "prompt": "Draft the claim notice to the supplier for CLM-5201, but don't send it.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=supplier-claim-engagement-draft-supplier-claim" }, { "label": "Claims Team Lead: Response deadline tracker", "prompt": "Which supplier claims are at risk of missing their response deadline?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=supplier-claim-engagement-sla-tracker" }, { "label": "Supplier Claims Coordinator: Supplier response evaluation", "prompt": "Litware Snacks replied on CLM-5204. What did they offer, and what's the gap?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=supplier-claim-engagement-supplier-response" }, { "label": "Claims Team Lead: Escalation brief draft", "prompt": "Prepare an escalation for the overdue shortage claim CLM-5202.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=supplier-claim-engagement-escalation-brief" } ], "architecture": { "business_flow": [ "Supplier claims coordinators and receiving teams", "Microsoft Teams or Microsoft 365 Copilot", "Supplier Claim Engagement Agent", "Approved order and receiving records", "Case system and supplier terms", "Authorized coordinator approval and send" ], "capabilities": [ { "operation": "claims_queue", "name": "Supplier claims queue", "purpose": "Use when a coordinator asks for open supplier claims, their value and severity, recovery to date, or what needs attention today." }, { "operation": "triage_claim", "name": "Claim triage", "purpose": "Use when a coordinator asks for a claim's severity, responsible supplier, purchase order, and supplier response and resolution windows." }, { "operation": "evidence_pack", "name": "Evidence pack check", "purpose": "Use when someone asks which photos and documents a claim needs, what is on file, and the claim value by line." }, { "operation": "draft_supplier_claim", "name": "Supplier claim notice draft", "purpose": "Use when a coordinator asks to prepare or draft the claim to a supplier; the notice stays Not Sent with the response dates it would carry." }, { "operation": "sla_tracker", "name": "Response deadline tracker", "purpose": "Use when the team asks which claims are breached, at risk, or on track against supplier response windows." }, { "operation": "supplier_response", "name": "Supplier response evaluation", "purpose": "Use when a coordinator asks what a supplier replied or offered and how it compares with the claim value." }, { "operation": "escalation_brief", "name": "Escalation brief draft", "purpose": "Use when the team asks to escalate a claim that is past its supplier response window; returns a draft for approval." } ], "easy_mode": [ "Verify Power Platform CLI 2.9.3 or newer and authenticate to the approved target environment.", "Initialize a modern cli-copilot project with an approved publisher prefix and preserve the generated identity settings.", "Upload the two packaged synthetic knowledge files and every operation-specific SKILL.md.", "Bind only approved least-privilege production connections after security, privacy, and business-owner review.", "Push the Draft agent, replay every locked persona case, and stop before publish for explicit approval." ], "local_install_prompt": "Install Supplier Claim Engagement Agent from https://raw.githubusercontent.com/microsoft/aibast-agents-library/main/solutions/supplier-claim-engagement/deployment.json, verify the expected tool and smoke case, and preserve the synthetic read-only boundary. Do not ask me to open a terminal. Do not connect customer systems or execute external actions.", "copilot_studio_prompt": "Using the Microsoft Copilot Studio plugin, initialize a Draft agent for Supplier Claim Engagement Agent from https://raw.githubusercontent.com/microsoft/aibast-agents-library/main/solutions/supplier-claim-engagement/deployment.json. Upload the two synthetic knowledge files and 7 skills, replay every locked case, and Stop before publish. Do not bind production connections without explicit customer approval.", "required_connections": [ "Use only approved, least-privilege production connections; the packaged demonstration uses synthetic knowledge files.", "Approved order and receiving records (read-only)", "Approved case system (read-only)", "Outlook send only through an approved, human-confirmed flow" ], "manual_commands": [ "pac auth create", "pac copilot init --name \"Supplier Claim Engagement Agent\" --publisher-prefix --authoring-mode cli-copilot --project-dir \"\" --environment \"\"", "pac connection list --environment \"\"", "pac copilot pull --project-dir \"\"", "pac copilot push --project-dir \"\"", "pac copilot publish --bot \"\" --environment \"\"" ], "acceptance_checks": [ "Every exposed operation has one persona-language case and one SKILL.md.", "Exactly two synthetic knowledge files are present.", "Responses state the relevant human approval or review boundary.", "No operation claims an external write, communication, commitment, or protected decision.", "Publishing remains a separate user-approved step." ], "hard_mode": [ "Run pac copilot init and verify settings.mcs.yml before authoring components.", "Author global instructions that preserve the packaged synthetic-only and human-approval boundaries.", "Add the two knowledge files and one skill for every exposed operation.", "Create dedicated connection references only for approved systems and verify least-privilege access.", "Pull before push, inspect every generated file, replay every locked case, and stop before publish." ] }, "curated_copy": true, "source": "field", "origin": "both", "onepager": "62. Supplier Claim Engagement Agent one-pager.pptx", "demo_video": "62-Supplier Claim Engagement Agent.mp4", "has_onepager": true, "has_demo_video": true, "promise_coverage": 100, "is_primary": true, "aliases": [], "package": { "slug": "supplier-claim-engagement", "quest_url": "solutions/supplier-claim-engagement/quest.html", "manual_tutorial_url": "solutions/supplier-claim-engagement/manual-tutorial.html", "export_manifest_url": "solutions/supplier-claim-engagement/export-manifest.json" } }, "_demo": { "slug": "supplier-claim-engagement", "case_count": 7, "personas": [ "Claims Team Lead", "Receiving Supervisor", "Supplier Claims Coordinator" ], "prompts": [ "Show me our open supplier claims and what needs attention today.", "Triage the new damage claim CLM-5201. Who is responsible and what are the terms?", "Is the evidence for CLM-5201 complete, and what is the claim worth?" ] }, "_catalog_kind": "solution", "_readiness": [ "Demo proven", "Conversation tested", "Synthetic data", "Production pattern" ] }, { "schema": "rapp-agent/1.0", "name": "@aibast-agents-library/supply-chain-disruption-alert", "version": "1.1.0", "display_name": "Supply Chain Disruption Alert Agent", "description": "Analyze a synthetic supply-chain snapshot for disruptions, route risk, mitigation scenarios, and alternative-supplier due diligence. Use for planning and continuity questions. The agent never changes a purchase order, activates a supplier, reroutes a shipment, or moves inventory; procurement and operations owners must approve actions through future authenticated tools.", "author": "AIBAST", "tags": [ "supply-chain", "disruption", "risk-management", "logistics", "retail" ], "category": "retail_cpg", "quality_tier": "verified", "requires_env": [], "dependencies": [ "@rapp/basic-agent" ], "_file": "agents/@aibast-agents-library/retail_cpg_stacks/supply_chain_disruption_alert_stack/supply_chain_disruption_alert_agent.py", "_size_kb": 41.5, "_lines": 1026, "_sha256": "6f7e8aac2c1b221a240c9bd9c65301f3207f382c038fcd1de21eb709935ffbdd", "_install_prefix": "supply_chain_disruption_alert__", "_install_filename": "supply_chain_disruption_alert__6f7e8aac2c1b_agent.py", "_stack": "supply_chain_disruption_alert", "_stack_vertical": "retail_cpg", "_synthetic_data": true, "_added_at": "2026-03-18T13:23:12-04:00", "_solution": { "advertised_name": "Supply Chain Disruption Alert Agent", "sharepoint_list_name": "Supply Chain Disruption Alert Agent", "slot": 35, "executive_summary": "Detects and manages supply chain risks to defend against disruptions, protect revenue, and maintain operational continuity.", "industries": [], "personas": [], "featured_tools": [ "Dynamics 365", "Microsoft Teams" ], "agent_requirements": [ "Microsoft Copilot Studio" ], "capabilities": [ "Detect supplier issues and logistics disruptions in real time", "Score severity and determine impacted SKUs, orders, and revenue exposure", "Compare backup sourcing, inventory shifts, and routing alternatives", "Recommend the best mitigation plan based on cost and lead time" ], "outcomes": [ "Acted early enough to reduce risk and protect commitments", "Boosted resilience by finding single\u2011source and geographic risks", "Activated alternate suppliers and shifted inventory to avoid delay", "Improved continuity and planning confidence during disruption events" ], "customer_scenario": [ "Production delays surfaced too late to prevent downstream impacts", "Supplier, logistics, and risk updates were scattered across systems", "Teams lacked clarity on which SKUs, orders, and customers were affected", "Mitigation decisions were reactive and often increased cost" ], "slide_title": "Supply Chain Disruption Alert Agent", "slide_summary": "Detects and manages supply chain risks to defend against disruptions, protect revenue, and maintain operational continuity.", "scenario_name": "A global electronic manufacturer", "challenge_intro": "An electronics manufacturer relied heavily on a concentrated supplier network for critical components.", "actions_intro": "The manufacturer used the Supply Chain Disruption Alert Agent to:", "outcomes_intro": "With the agent, the company:", "opportunity_statements": [ "Detects and manages supply chain risks to defend against disruptions, protect revenue, and maintain operational continuity.", "Monitor supplier, logistics, and market signals to detect early signs of disruption and trigger rapid response workflows", "Analyze impacts on products, orders, and revenue to prioritize mitigation and guide effective decision\u2011making", "Recommend alternative sourcing, routing, and inventory actions to reduce operational risk and main supply continuity" ], "module_name": null, "module_description": null, "source_verified": true, "sales_headline": "See disruption trade-offs before anyone changes the network", "card_pitch": "Turn synthetic disruption events, route risk, mitigation costs, and supplier candidates into a shared continuity brief while keeping sourcing, ordering, routing, communication, and inventory movement under human control.", "why_try": "A supply chain planner can ask what is at risk and receive the exact route, event, SKU, cost, alternative, and approval boundary.", "customer_challenge_copy": "Supplier, logistics, order, and external-risk signals are scattered, so disruption impacts surface late and reactive mitigation can increase cost or create new operational risk.", "microsoft_ai_story": "Microsoft Copilot Studio can unify governed review across Dynamics 365 Supply Chain Management, approved supplier-risk feeds, transportation management, Teams, and controlled procurement workflows.", "business_value": [ "Improves visibility into disruption severity and modeled exposure.", "Supports consistent comparison of mitigation and supplier alternatives.", "Preserves procurement and inventory authority by separating recommendations from network execution." ], "search_terms": [ "supply chain disruption alert", "supply chain risk", "route risk", "supplier alternatives", "mitigation planning", "revenue exposure", "Dynamics 365 Supply Chain Management", "transportation management", "procurement approval", "inventory movement", "Microsoft Teams", "Copilot Studio" ], "journey_stage": "Optimize", "blueprint_role": "Creates a governed continuity-planning layer before supplier activation, purchase-order changes, shipment rerouting, or inventory movement.", "sample_prompts": [ { "label": "Customer Fulfillment Lead: Active disruption brief", "prompt": "Which active disruption has the largest modeled impact and what is affected?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=supply-chain-disruption-alert-disruption-dashboard" }, { "label": "Supply Chain Planner: Route risk explanation", "prompt": "Why is RT-APAC-01 high risk?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=supply-chain-disruption-alert-risk-assessment" }, { "label": "Operations Leader: Approval-gated mitigation scenario", "prompt": "Draft a DISR-002 mitigation scenario without rerouting or moving inventory.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=supply-chain-disruption-alert-mitigation-plan" }, { "label": "Procurement Manager: Supplier due-diligence candidates", "prompt": "Show Electronics alternatives without activating a supplier.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=supply-chain-disruption-alert-supplier-alternatives" }, { "label": "Regional Retail Manager: Northwest root-cause analysis", "prompt": "I'm seeing unusual inventory movement at our Northwest stores. Can you help me understand what's happening?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=supply-chain-disruption-alert-disruption-dashboard" }, { "label": "Regional Operations Director: Emergency response options", "prompt": "What are our emergency options and costs?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=supply-chain-disruption-alert-disruption-dashboard" }, { "label": "Store Operations Manager: Ready-to-release transfer plan", "prompt": "Approved. Execute both Seattle and the 5 additional stores.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=supply-chain-disruption-alert-disruption-dashboard" }, { "label": "Distribution Center Manager: DC recovery plan", "prompt": "Activate tracking and show me the Portland DC recovery plan.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=supply-chain-disruption-alert-disruption-dashboard" }, { "label": "Finance Business Partner: Executive incident report", "prompt": "Create the executive report with financial impact.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=supply-chain-disruption-alert-disruption-dashboard" }, { "label": "Supply Chain Director: Crisis response summary", "prompt": "Distribute the report and summarize what we accomplished.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=supply-chain-disruption-alert-disruption-dashboard" } ], "architecture": { "business_flow": [ "Supply chain planners, operations leaders, procurement managers, and customer fulfillment leads", "Microsoft 365 Copilot or Copilot Studio", "Supply Chain Disruption Alert Agent", "Dynamics 365 Supply Chain Management", "Approved supplier-risk and logistics feeds", "Transportation management and Microsoft Teams", "Authorized procurement and operations owners" ], "capabilities": [ { "operation": "disruption_dashboard", "name": "Active disruption brief", "purpose": "Summarizes synthetic events, routes, SKUs, delay, and modeled exposure." }, { "operation": "risk_assessment", "name": "Route risk explanation", "purpose": "Explains synthetic geopolitical, weather, infrastructure, labor, regulatory, and financial factors." }, { "operation": "mitigation_plan", "name": "Approval-gated mitigation scenario", "purpose": "Compares response actions and costs without changing a purchase order, shipment, supplier, or inventory position." }, { "operation": "supplier_alternatives", "name": "Supplier due-diligence candidates", "purpose": "Presents synthetic supplier evidence without qualification, contracting, activation, or ordering." }, { "operation": "root_cause_analysis", "name": "Northwest root-cause analysis", "purpose": "Explains the Portland DC disruption: issue, impact and status table, affected categories and customer impact." }, { "operation": "emergency_options", "name": "Emergency response options", "purpose": "Compares the Denver DC emergency transfer with partial fill and wait by timeline, coverage, cost, recovery and ROI." }, { "operation": "transfer_plan", "name": "Ready-to-release transfer plan", "purpose": "Prepares the Seattle and five-store transfer plan and coordination drafts without dispatching anything." }, { "operation": "recovery_plan", "name": "DC recovery plan", "purpose": "Shows the shipment tracking snapshot, Portland DC recovery actions, backlog clearance and prevention recommendation." }, { "operation": "incident_report", "name": "Executive incident report", "purpose": "Drafts financial impact, response performance, lessons learned and prevention value for leadership review." }, { "operation": "incident_summary", "name": "Crisis response summary", "purpose": "Summarizes the response end to end as a draft package ready for the user to distribute." } ], "easy_mode": [ "Verify Power Platform CLI 2.9.3 or newer and authenticate to the target environment.", "Use the Copilot Studio plugin initialization workflow to create a modern, sync-connected Draft agent with an approved publisher prefix.", "Upload manual/GLOBAL-INSTRUCTIONS.md, the two synthetic knowledge files, and all 4 operation skills from the solution package.", "Bind only approved, least-privilege production connections through dedicated connection references and preserve the human approval boundary.", "Push the Draft project, replay every locked persona prompt, inspect safety behavior, and stop before publishing." ], "local_install_prompt": "You are GitHub Copilot running in Agent mode in VS Code. Install and open the AI BAST Supply Chain Disruption Alert Agent from this deployment recipe: https://raw.githubusercontent.com/microsoft/aibast-agents-library/main/solutions/supply-chain-disruption-alert/deployment.json. Own the setup and do not ask me to open a terminal. Verify the Brainstem health endpoint, install or start it if needed using the official recipe, fetch the registry and source, verify the source hash, place the exact agent file in the discovered agents directory, confirm the expected tool, run the smoke prompt, and open the local UI. Preserve the synthetic read-only boundary. Do not connect customer systems, perform an external action, publish, or silently substitute another agent. Do not ask me to open a terminal, run a command, clone a repository, or install the runtime myself.", "copilot_studio_prompt": "You are GitHub Copilot running in Agent mode in VS Code with the Microsoft Copilot Studio plugin installed. Promote the AI BAST Supply Chain Disruption Alert Agent using this deployment recipe: https://raw.githubusercontent.com/microsoft/aibast-agents-library/main/solutions/supply-chain-disruption-alert/deployment.json. Ask only for the target environment ID and publisher prefix. Initialize a modern cli-copilot Draft project, upload the packaged global instructions, two synthetic knowledge files, and 4 skills, and preserve these operations: disruption_dashboard, risk_assessment, mitigation_plan, supplier_alternatives. Bind only approved dedicated connections, review every generated file and connection, pull before push, replay every locked persona case, enforce explicit authorized approval for any future write, and stop before publish. Show validation evidence and unresolved connection work. Stop before publish.", "required_connections": [ "Dynamics 365 Supply Chain Management", "Approved supplier-risk and market feed", "Transportation management system", "Procurement and contract workflow", "Microsoft Teams for authorized collaboration" ], "manual_commands": [ "pac auth create", "pac copilot init --name \"Supply Chain Disruption Alert Agent\" --publisher-prefix --authoring-mode cli-copilot --project-dir \"\" --environment \"\"", "pac connection list --environment \"\"", "pac copilot pull --project-dir \"\"", "pac copilot push --project-dir \"\"", "pac copilot publish --bot \"\" --environment \"\"" ], "acceptance_checks": [ "settings.mcs.yml preserves the initialized identity, environment, language, and authoring mode.", "All 10 implemented operations are represented by instructions, a packaged skill, or a bound tool.", "Exactly two synthetic knowledge files are loaded and clearly labeled fictional.", "Every production source uses a dedicated, least-privilege connection reference.", "Every locked persona prompt returns its deterministic synthetic evidence and the relevant human approval boundary.", "Never activate or contact a supplier, change a purchase order, or reroute a shipment.", "Never move inventory or claim a mitigation action occurred.", "Public value statements remain qualitative; exact figures appear only as synthetic evidence.", "Publishing remains a separate, user-approved step after push and validation." ], "hard_mode": [ "Run pac copilot init and confirm settings.mcs.yml exists.", "Author global instructions that route every operation and preserve the packaged no-write boundaries.", "Upload the two synthetic Markdown knowledge files and 4 operation-specific SKILL.md files.", "Create read tools first; isolate any future write action behind authorization, current-state validation, explicit confirmation, and immutable audit logging.", "Validate every component schema and filename, push the Draft, replay every locked prompt, inspect failures, and stop before publish." ] }, "curated_copy": true, "source": "catalog", "origin": null, "onepager": "35. Supply Chain Disruption Alert Agent one-pager.pptx", "demo_video": null, "has_onepager": true, "has_demo_video": false, "promise_coverage": 85, "is_primary": true, "aliases": [ "Supply Risk Monitoring Agent" ], "package": { "slug": "supply-chain-disruption-alert", "quest_url": "solutions/supply-chain-disruption-alert/quest.html", "manual_tutorial_url": "solutions/supply-chain-disruption-alert/manual-tutorial.html", "export_manifest_url": "solutions/supply-chain-disruption-alert/export-manifest.json" } }, "_demo": { "slug": "supply-chain-disruption-alert", "case_count": 10, "personas": [ "Customer Fulfillment Lead", "Distribution Center Manager", "Finance Business Partner", "Operations Leader", "Procurement Manager", "Regional Operations Director", "Regional Retail Manager", "Store Operations Manager", "Supply Chain Director", "Supply Chain Planner" ], "prompts": [ "Which active disruption has the largest modeled impact and what is affected?", "Why is RT-APAC-01 high risk?", "Draft a DISR-002 mitigation scenario without rerouting or moving inventory." ] }, "_catalog_kind": "solution", "_readiness": [ "Demo proven", "Conversation tested", "Synthetic data", "Production pattern" ] }, { "schema": "rapp-agent/1.0", "name": "@aibast-agents-library/briefing-pack-builder", "version": "1.0.0", "display_name": "Briefing Pack Builder Agent", "description": "Turn a government team's source documents into cited, template-ready briefs in minutes, with conflicting figures flagged and missing content named, so specialists verify instead of rebuild.", "author": "AIBAST", "tags": [ "government", "state-and-local", "briefing", "budget-hearing", "document-drafting" ], "category": "slg_government", "quality_tier": "verified", "requires_env": [], "dependencies": [ "@rapp/basic-agent" ], "_file": "agents/@aibast-agents-library/slg_government_stacks/briefing_pack_builder_stack/briefing_pack_builder_agent.py", "_size_kb": 20.4, "_lines": 392, "_sha256": "66db05eb81c8f9ccb170ccd3a501a77a49672f96fb126025bec1a63c8ddac4b3", "_install_prefix": "briefing_pack_builder__", "_install_filename": "briefing_pack_builder__66db05eb81c8_agent.py", "_stack": "briefing_pack_builder", "_stack_vertical": "slg_government", "_synthetic_data": true, "_added_at": "2026-10-08T10:30:56-04:00", "_solution": { "advertised_name": "Briefing Pack Builder Agent", "sharepoint_list_name": "Briefing Pack Builder Agent", "slot": 66, "executive_summary": "Turn a government team's source documents into cited, template-ready briefs in minutes, with conflicting figures flagged and missing content named, so specialists verify instead of rebuild.", "industries": [ "State and Local Government" ], "personas": [ "Briefing Coordinators", "Budget Analysts", "Department Directors" ], "featured_tools": [ "Microsoft Word", "SharePoint", "Microsoft Teams" ], "agent_requirements": [ "Microsoft Copilot Studio" ], "capabilities": [ "Lists formats and scans what the content set holds", "Flags figures that disagree across documents", "Drafts the brief with a citation for every section", "Names gaps and the review and approval route" ], "outcomes": [ "Faster first drafts that specialists verify, not rebuild", "Fewer wrong figures in front of decision makers", "Earlier requests for missing content" ], "customer_scenario": [ "Briefs are rebuilt by hand for every format", "Superseded figures slip into briefs", "Missing content surfaces only at review", "Sign-off routes and timelines are unclear" ], "slide_title": "Briefing Pack Builder Agent", "slide_summary": "Turns a government team's source documents into cited, template-ready briefs in minutes, flagging conflicting figures and naming missing content.", "scenario_name": "A city budget office", "challenge_intro": "Government briefing teams build high-stakes briefs under tight deadlines:", "actions_intro": "The Briefing Pack Builder Agent drafts from governed content:", "outcomes_intro": "Briefing teams that adopt the agent can expect:", "opportunity_statements": [ "Draft briefs from the documents you already hold", "Catch conflicting figures before a hearing", "Name missing content on day one", "Know the sign-off route for every format" ], "module_name": null, "module_description": null, "source_verified": true, "sales_headline": "Draft the brief from the documents, and show every source", "card_pitch": "Give government briefing teams cited, template-ready drafts from one governed document set, with conflicting figures listed for a person to settle and missing sections named, while review and sign-off stay with officers.", "why_try": "The demo takes a briefing coordinator from the list of formats to a cited budget hearing brief, a conflicting-figure check, a gap report and the sign-off route in six questions.", "customer_challenge_copy": "Briefing teams rebuild every brief by hand from budget papers, status reports and transcripts, quote figures that have since changed, and discover missing content only when a reviewer sends the draft back.", "microsoft_ai_story": "Copilot Studio gives briefing teams a Teams or Microsoft 365 Copilot drafting assistant over an approved SharePoint document library; drafts cite their sources, figures that disagree are listed for confirmation, and approval stays a human sign-off.", "business_value": [ "Briefs drafted in minutes with a source behind every statement.", "Superseded figures caught before they reach a hearing.", "Missing content requested early instead of at review." ], "search_terms": [ "Briefing Coordinator", "Briefing Pack Builder Agent", "Budget Analyst", "Department Director", "Microsoft Copilot Studio", "Microsoft Teams", "Microsoft Word", "SharePoint", "approval_route", "content_scan", "draft_brief", "figure_check", "gap_report", "list_templates" ], "journey_stage": "Create", "blueprint_role": "Adds a cited drafting layer over a governed document library, separating draft preparation from figure confirmation, review and sign-off.", "sample_prompts": [ { "label": "Briefing Coordinator: Briefing formats", "prompt": "What briefing formats can you build for us?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=briefing-pack-builder-list-templates" }, { "label": "Budget Analyst: Content scan", "prompt": "What do we hold on the Elm Street Bridge Replacement?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=briefing-pack-builder-content-scan" }, { "label": "Budget Analyst: Figure check", "prompt": "Before I draft anything, do any of the figures disagree?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=briefing-pack-builder-figure-check" }, { "label": "Briefing Coordinator: Cited draft brief", "prompt": "Draft the budget hearing brief on the Elm Street bridge.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=briefing-pack-builder-draft-brief" }, { "label": "Briefing Coordinator: Gap report", "prompt": "Which sections of that brief still need input?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=briefing-pack-builder-gap-report" }, { "label": "Department Director: Approval route", "prompt": "Who signs this brief off, and how long does it take?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=briefing-pack-builder-approval-route" } ], "architecture": { "business_flow": [ "Briefing coordinators and budget analysts", "Microsoft Teams or Microsoft 365 Copilot", "Briefing Pack Builder Agent", "Approved SharePoint document library", "Owning department (figures and gaps)", "Reviewer and approver sign-off (human decision)" ], "capabilities": [ { "operation": "list_templates", "name": "Briefing formats", "purpose": "Use when someone asks which briefing formats or templates are available." }, { "operation": "content_scan", "name": "Content scan", "purpose": "Use when someone asks what the content set holds or has on a topic." }, { "operation": "figure_check", "name": "Figure check", "purpose": "Use when someone asks whether figures disagree or conflict across documents on a topic." }, { "operation": "draft_brief", "name": "Cited draft brief", "purpose": "Use when someone asks to draft, prepare or build a brief in a chosen format on a topic." }, { "operation": "gap_report", "name": "Gap report", "purpose": "Use when someone asks which sections of a brief are missing content or still need input." }, { "operation": "approval_route", "name": "Approval route", "purpose": "Use when someone asks who reviews or signs off a brief and how long approval takes." } ], "easy_mode": [ "Verify Power Platform CLI 2.9.3 or newer and authenticate to the approved target environment.", "Initialize a modern cli-copilot project with an approved publisher prefix and preserve the generated identity settings.", "Upload the two packaged synthetic knowledge files and every operation-specific SKILL.md.", "Bind only approved least-privilege production connections after security, privacy, and business-owner review.", "Push the Draft agent, replay every locked persona case, and stop before publish for explicit approval." ], "local_install_prompt": "Install Briefing Pack Builder Agent from https://raw.githubusercontent.com/microsoft/aibast-agents-library/main/solutions/briefing-pack-builder/deployment.json, verify the expected tool and smoke case, and preserve the synthetic read-only boundary. Do not ask me to open a terminal. Do not connect customer systems or execute external actions.", "copilot_studio_prompt": "Using the Microsoft Copilot Studio plugin, initialize a Draft agent for Briefing Pack Builder Agent from https://raw.githubusercontent.com/microsoft/aibast-agents-library/main/solutions/briefing-pack-builder/deployment.json. Upload the two synthetic knowledge files and 6 skills, replay every locked case, and Stop before publish. Do not bind production connections without explicit customer approval.", "required_connections": [ "Use only approved, least-privilege production connections; the packaged demonstration uses synthetic knowledge files.", "Approved SharePoint document library (read-only)", "Approved drafts folder for officer-saved briefs (no agent write in the demonstration)" ], "manual_commands": [ "pac auth create", "pac copilot init --name \"Briefing Pack Builder Agent\" --publisher-prefix --authoring-mode cli-copilot --project-dir \"\" --environment \"\"", "pac connection list --environment \"\"", "pac copilot pull --project-dir \"\"", "pac copilot push --project-dir \"\"", "pac copilot publish --bot \"\" --environment \"\"" ], "acceptance_checks": [ "Every exposed operation has one persona-language case and one SKILL.md.", "Exactly two synthetic knowledge files are present.", "Responses state the relevant human approval or review boundary.", "No operation claims an external write, communication, commitment, or protected decision.", "Publishing remains a separate user-approved step." ], "hard_mode": [ "Run pac copilot init and verify settings.mcs.yml before authoring components.", "Author global instructions that preserve the packaged synthetic-only and human-approval boundaries.", "Add the two knowledge files and one skill for every exposed operation.", "Create dedicated connection references only for approved systems and verify least-privilege access.", "Pull before push, inspect every generated file, replay every locked case, and stop before publish." ] }, "curated_copy": true, "source": "field", "origin": "both", "onepager": "66. Briefing Pack Builder Agent one-pager.pptx", "demo_video": "66-Briefing Pack Builder Agent.mp4", "has_onepager": true, "has_demo_video": true, "promise_coverage": 100, "is_primary": true, "aliases": [], "package": { "slug": "briefing-pack-builder", "quest_url": "solutions/briefing-pack-builder/quest.html", "manual_tutorial_url": "solutions/briefing-pack-builder/manual-tutorial.html", "export_manifest_url": "solutions/briefing-pack-builder/export-manifest.json" } }, "_demo": { "slug": "briefing-pack-builder", "case_count": 6, "personas": [ "Briefing Coordinator", "Budget Analyst", "Department Director" ], "prompts": [ "What briefing formats can you build for us?", "What do we hold on the Elm Street Bridge Replacement?", "Before I draft anything, do any of the figures disagree?" ] }, "_catalog_kind": "solution", "_readiness": [ "Demo proven", "Conversation tested", "Synthetic data", "Production pattern" ] }, { "schema": "rapp-agent/1.0", "name": "@aibast-agents-library/building-permit-processing", "version": "1.1.0", "display_name": "Building Permit Processing Agent", "description": "Automate building permit review processes to enable faster service, lower operational costs, and higher citizen satisfaction.", "author": "AIBAST", "tags": [ "permits", "building", "zoning", "inspection", "local-government", "fees" ], "category": "slg_government", "quality_tier": "verified", "requires_env": [], "dependencies": [ "@rapp/basic-agent" ], "_file": "agents/@aibast-agents-library/slg_government_stacks/building_permit_processing_stack/building_permit_processing_agent.py", "_size_kb": 44.9, "_lines": 945, "_sha256": "f263fce15e46a25788db493139e6928c41815e8d8187f53335e71e705365b57d", "_install_prefix": "building_permit_processing__", "_install_filename": "building_permit_processing__f263fce15e46_agent.py", "_stack": "building_permit_processing", "_stack_vertical": "slg_government", "_synthetic_data": true, "_added_at": "2026-03-18T13:23:12-04:00", "_solution": { "advertised_name": "Building Permit Processing Agent", "sharepoint_list_name": "Building Permit Review Agent", "slot": 44, "executive_summary": "Automate building permit review processes to enable faster service, lower operational costs, and higher citizen satisfaction.", "industries": [ "State and Local Government" ], "personas": [ "Permit Technicians", "Plan Reviewers", "Inspectors" ], "featured_tools": [ "Dynamics 365 Customer Service", "Microsoft Teams", "SharePoint" ], "agent_requirements": [ "Microsoft Copilot Studio" ], "capabilities": [ "Classify and validate permit applications at intake", "Route requests automatically to the correct review teams", "Centralize communication and provide instant updates to applicants", "Provide inspectors with mobile access and real\u2011time assignments" ], "outcomes": [ "Accelerated approvals with consistent, automated workflows", "Reduced status\u2011check calls through transparent, self\u2011service tracking", "Improved coordination across planning, zoning, code, and inspections", "Delivered a faster, more predictable permitting experience for residents" ], "customer_scenario": [ "Manual routing caused delays and inconsistent approvals", "Applicants repeatedly called for status updates", "Duplicate or incomplete applications created rework", "High staff workloads slowed permit\u2011to\u2011inspection timelines" ], "slide_title": "Building Permit Processing Agent", "slide_summary": "Automate building permit review processes to enable faster service, lower operational costs, and higher citizen satisfaction.", "scenario_name": "A city development services department", "challenge_intro": "A city\u2019s development services department faced long processing times and high citizen complaints.", "actions_intro": "The department used the Building Permit Processing Agent to:", "outcomes_intro": "With the agent, the department:", "opportunity_statements": [ "Automate permit intake and classification to ensure completeness and reduce manual review time", "Route applications intelligently across departments to prevent bottlenecks and achieve more timely reviews", "Provide real\u2011time tracking and proactive updates to reduce inquiries and improve service quality" ], "module_name": null, "module_description": null, "source_verified": true, "sales_headline": "Turn permit backlogs into a predictable citizen service", "card_pitch": "Give development services teams one place to spot late reviews, reject bad intake, route new applications, prepare applicant updates, and coordinate inspections before delays turn into complaints.", "why_try": "This demo makes public-sector AI immediately tangible: ask one plain-language question and expose the application, owner, review desk, or inspector that needs action next.", "customer_challenge_copy": "A city development services department is balancing long review cycles, duplicate or incomplete submissions, repeated status calls, and inspectors who need current assignments without chasing disconnected systems.", "microsoft_ai_story": "Microsoft Copilot Studio is the conversational front door. Dynamics 365 Customer Service manages the permit case, SharePoint supplies plans and supporting documents, and Microsoft Teams brings assignments and updates to reviewers and inspectors.", "business_value": [ "Improves intake consistency and reduces avoidable rework before review begins.", "Accelerates routing and gives staff a clearer view of overdue and at-risk applications.", "Strengthens applicant communication and coordination across planning, zoning, code review, and inspections." ], "search_terms": [ "building permit backlog", "development services", "permit intake", "plan review", "zoning review", "inspection scheduling", "citizen status calls", "Dynamics 365 Customer Service", "SharePoint permit documents", "Microsoft Teams inspectors", "local government", "Copilot Studio" ], "journey_stage": "Govern", "blueprint_role": "Establishes a governed public-service workflow where intake, review clocks, applicant communication, and field inspections share one conversational control point.", "sample_prompts": [ { "label": "Find the permit most likely to trigger a complaint", "prompt": "Which permit applications have been sitting too long, and which resident is going to complain first?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=permit-backlog" }, { "label": "Triage today\u2019s front-counter submissions", "prompt": "Anything at the front counter I should not be accepting today?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=permit-intake" }, { "label": "Route the Harbor Way restaurant fit-out", "prompt": "The restaurant fit-out on Harbor Way just came in. Who needs to review it and when is it due back?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=permit-routing" }, { "label": "Stop avoidable applicant status calls", "prompt": "My front desk is drowning in status calls. What can we send out today so people stop calling?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=permit-updates" }, { "label": "See the solar inspection board", "prompt": "What inspections are on the board for the solar job, and who is covering them?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=permit-inspections" } ], "architecture": { "business_flow": [ "Permit applicants and city staff", "Microsoft 365 Copilot or Copilot Studio", "Building Permit Processing Agent", "Dynamics 365 Customer Service", "SharePoint", "Microsoft Teams", "Reviewers and inspectors" ], "capabilities": [ { "operation": "permit_backlog", "name": "Backlog and complaint-risk view", "purpose": "Ranks open applications against review clocks and correction cycles so managers know where intervention matters." }, { "operation": "intake_triage", "name": "Intake validation and routing", "purpose": "Detects duplicates and missing documents, classifies the permit, and identifies the review desks and due date." }, { "operation": "applicant_updates", "name": "Proactive applicant communication", "purpose": "Drafts clear status updates from the current permit state so front-desk teams can reduce avoidable status calls." }, { "operation": "inspector_assignment", "name": "Inspection coordination", "purpose": "Surfaces scheduled inspections, inspector specialties, availability, and the job context needed for assignment." } ], "easy_mode": [ "Verify Power Platform CLI 2.9.3 or newer and authenticate to the target environment.", "Use the Copilot Studio plugin initialization workflow to create a modern, sync-connected agent project with an approved publisher prefix.", "Give the Copilot Studio Architect this blueprint and the initialized project so it writes the modern instructions, skills, and tools as YAML.", "Bind Dynamics 365 Customer Service, SharePoint, and Microsoft Teams connections through dedicated connection references.", "Use the Copilot Studio Manage workflow to run pac copilot push, then validate the three canned prompts before publishing." ], "local_install_prompt": "You are GitHub Copilot running in Agent mode in VS Code. Install and open the AI BAST Building Permit Processing Agent from this deployment recipe: https://raw.githubusercontent.com/microsoft/aibast-agents-library/main/solutions/building-permit-processing/deployment.json. Own the entire setup. Do not ask me to open a terminal, run a command, clone a repository, or install the Brainstem myself. Inspect the operating system; check the recipe health endpoint; if the Brainstem is missing, run the official installer yourself using the recipe's OS-appropriate installer; install or refresh the policy-clean launcher; fetch registry.json and the raw agent source; verify the file against the registry SHA-256; place it under the discovered Brainstem agents directory with the recipe target filename; start the Brainstem; verify /health reports the expected tool; run the recipe smoke prompt through /chat and confirm the expected agent fired; then open the local chat UI in my browser. If authentication requires a browser confirmation, launch the supported device flow and ask me only to complete that browser step. Do not publish, connect customer systems, or silently substitute another agent. If any step fails, stop and show the exact failed step and output.", "copilot_studio_prompt": "You are GitHub Copilot running in Agent mode in VS Code with the Microsoft Copilot Studio plugin installed. Promote the AI BAST Building Permit Processing Agent using this deployment recipe: https://raw.githubusercontent.com/microsoft/aibast-agents-library/main/solutions/building-permit-processing/deployment.json. Own every command and file change; do not ask me to use a terminal. Ask me only for the target Power Platform environment ID and publisher prefix. Verify PAC CLI is at least 2.9.3, authenticate if needed, initialize a new cli-copilot project, and use the Copilot Studio Architect workflow to author modern YAML that preserves permit_backlog, intake_triage, applicant_updates, and inspector_assignment. Create or bind dedicated Dynamics 365 Customer Service, SharePoint, and Microsoft Teams connection references, review every generated YAML file and connection, pull before push, run pac copilot push, and replay the three recipe acceptance prompts. Stop before publish and show me the generated files, validation evidence, unresolved connection work, and the exact publish command that would make it live.", "required_connections": [ "Dynamics 365 Customer Service", "SharePoint", "Microsoft Teams" ], "manual_commands": [ "pac auth create", "pac copilot init --name \"Building Permit Processing Agent\" --publisher-prefix --authoring-mode cli-copilot --project-dir \"\" --environment \"\"", "pac connection list --environment \"\"", "pac copilot pull --project-dir \"\"", "pac copilot push --project-dir \"\"", "pac copilot publish --bot \"\" --environment \"\"" ], "acceptance_checks": [ "settings.mcs.yml exists and preserves the initialized identity, environment binding, language, and authoring mode.", "Every advertised permit capability is represented by instructions, a skill, or a bound tool.", "Connection references point to dedicated, connected records in the target environment.", "The backlog, intake, routing, communication, and inspection demo prompts produce the expected entities and actions.", "Publishing remains a separate, user-approved step after push and validation." ], "hard_mode": [ "Run pac copilot init and confirm settings.mcs.yml exists.", "Author global instructions for intake, statutory-clock awareness, routing, applicant communication, approvals, and safe escalation.", "Create skills or tools that map the backlog, intake, update, and inspection operations to production connectors or workflows.", "Create and verify dedicated Dataverse connection references for Dynamics 365 Customer Service, SharePoint, and Microsoft Teams.", "Validate every component filename and schema, run pac copilot push, inspect the generated agent in Copilot Studio, and replay the canned prompts." ] }, "curated_copy": true, "source": "catalog", "origin": null, "onepager": "46. Building Permit Processing Agent one-pager.pptx", "demo_video": "46-Building Permit Processing Agent.mp4", "has_onepager": true, "has_demo_video": true, "promise_coverage": 40, "is_primary": true, "aliases": [], "package": { "slug": "building-permit-processing", "quest_url": "solutions/building-permit-processing/quest.html", "manual_tutorial_url": "solutions/building-permit-processing/manual-tutorial.html", "export_manifest_url": "solutions/building-permit-processing/export-manifest.json" } }, "_demo": { "slug": "building-permit-processing", "case_count": 10, "personas": [ "Chief Building Inspector", "Customer Service Lead", "Development Services Director", "Permit Technician", "Permitting Supervisor" ], "prompts": [ "Which permit applications have been sitting too long, and which resident is going to complain first?", "Anything at the front counter I should not be accepting today?", "The restaurant fit-out on Harbor Way just came in. Who needs to review it and when is it due back?" ] }, "_catalog_kind": "solution", "_readiness": [ "Demo proven", "Conversation tested", "Synthetic data", "Production pattern" ] }, { "schema": "rapp-agent/1.0", "name": "@aibast-agents-library/public-correspondence-response", "version": "1.0.0", "display_name": "Public Correspondence Agent", "description": "Answer every part of every resident enquiry from approved precedent: triage the correspondence queue, break multi-part letters into separate questions, cite approved past responses, route unanswered points to the owning team, and prepare a reviewable draft reply.", "author": "AIBAST", "tags": [ "government", "correspondence", "constituent-services", "public-works", "drafting", "knowledge-reuse" ], "category": "slg_government", "quality_tier": "verified", "requires_env": [], "dependencies": [ "@rapp/basic-agent" ], "_file": "agents/@aibast-agents-library/slg_government_stacks/public_correspondence_response_stack/public_correspondence_response_agent.py", "_size_kb": 24.7, "_lines": 461, "_sha256": "45564b5071ed5c0d1577e8a82d49b969f2d96a23615d505bbbaa29d4c2b7d09e", "_install_prefix": "public_correspondence_response__", "_install_filename": "public_correspondence_response__45564b5071ed_agent.py", "_stack": "public_correspondence_response", "_stack_vertical": "slg_government", "_synthetic_data": true, "_added_at": "2026-10-08T10:30:56-04:00", "_solution": { "advertised_name": "Public Correspondence Agent", "sharepoint_list_name": "Public Correspondence Agent", "slot": 58, "executive_summary": "Answer every part of every resident enquiry from approved precedent: triage the correspondence queue, break multi-part letters into separate questions, cite approved past responses, route unanswered points to the owning team, and prepare a reviewable draft reply.", "industries": [ "State and Local Government" ], "personas": [ "Correspondence Officer", "Correspondence Team Lead", "Knowledge Owner" ], "featured_tools": [ "Microsoft Outlook", "Microsoft SharePoint", "Microsoft Teams" ], "agent_requirements": [ "Microsoft Copilot Studio" ], "capabilities": [ "Triages letters, emails, forms and phone notes into one queue", "Splits multi-part enquiries so every question is answered", "Finds approved past responses for each question and cites them", "Routes unanswered points to the owning team with the question" ], "outcomes": [ "Every question in every letter is addressed in the draft", "Approved wording is reused with a citation instead of rewritten", "Gaps reach the right team in minutes and become precedent" ], "customer_scenario": [ "Parts of multi-part letters are missed in replies", "Approved wording is hard to find and gets rewritten", "New questions stall while ownership is worked out", "New approved answers are not captured for reuse" ], "slide_title": "Public Correspondence Agent", "slide_summary": "Answers every part of every resident enquiry from approved precedent, with cited drafts and gaps routed to the owning team.", "scenario_name": "A county public works team", "challenge_intro": "Correspondence teams face rising volumes of multi-part enquiries:", "actions_intro": "The Public Correspondence Agent gives officers a cited, complete draft:", "outcomes_intro": "Correspondence teams respond completely and consistently:", "opportunity_statements": [ "Every resident question answered, not just the first one", "Approved wording reused with citations", "New questions routed to the right team in minutes", "Officers keep sign-off on every reply" ], "module_name": null, "module_description": null, "source_verified": true, "sales_headline": "Answer every part of every resident letter from approved wording", "card_pitch": "Give correspondence officers one queue for letters, emails, web forms and phone notes, a breakdown of every separate question, cited approved answers, and a clear referral for anything new, while officers stay in control of every reply.", "why_try": "The demo walks one multi-part resident letter from the queue to a cited draft reply, with the unanswered question routed to the right team and nothing sent.", "customer_challenge_copy": "Correspondence teams answer high volumes of multi-part resident enquiries across many channels; parts get missed, approved wording is rewritten, and new questions stall while ownership is worked out.", "microsoft_ai_story": "Copilot Studio and Microsoft Teams give officers the assistant; approved correspondence records, a SharePoint library of approved responses and the team directory can be bound with least-privilege access, while Outlook stays the officer's own sending channel.", "business_value": [ "Every question in a resident enquiry is addressed before a reply goes out.", "Approved wording is reused with a citation instead of rewritten.", "New questions reach the owning team quickly and become reusable precedent." ], "search_terms": [ "Correspondence Officer", "Correspondence Team Lead", "Knowledge Owner", "Microsoft Copilot Studio", "Microsoft Outlook", "Microsoft SharePoint", "Microsoft Teams", "Public Correspondence Agent", "break_down_enquiry", "draft_response", "find_precedents", "inbox_queue", "library_update", "route_and_refer" ], "journey_stage": "Serve", "blueprint_role": "Separates correspondence triage, precedent retrieval and draft preparation from the officer's sending and sign-off authority.", "sample_prompts": [ { "label": "Correspondence Officer: Correspondence queue triage", "prompt": "What correspondence do I need to answer this week?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=public-correspondence-response-inbox-queue" }, { "label": "Correspondence Officer: Enquiry breakdown", "prompt": "Break down the scanned letter from E. Okafor into its separate questions.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=public-correspondence-response-break-down-enquiry" }, { "label": "Correspondence Officer: Approved precedent match", "prompt": "Which approved past responses already answer each question in the Okafor letter?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=public-correspondence-response-find-precedents" }, { "label": "Correspondence Team Lead: Team routing and referral", "prompt": "Who owns each part of the Okafor letter, and what should I ask the team about the part we can't answer?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=public-correspondence-response-route-and-refer" }, { "label": "Correspondence Officer: Cited draft reply", "prompt": "Draft the reply to the Okafor letter for my review.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=public-correspondence-response-draft-response" }, { "label": "Knowledge Owner: Precedent library update", "prompt": "Property and Land will send approved wording for the Okafor compensation question. How do we add it to the precedent library so nobody has to ask again?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=public-correspondence-response-library-update" } ], "architecture": { "business_flow": [ "Residents writing by letter, email, web form or phone", "Correspondence officers in Microsoft Teams or Microsoft 365 Copilot", "Public Correspondence Agent", "Approved correspondence records and response library", "Owning teams for new questions", "Officer review, sign-off and sending in Outlook" ], "capabilities": [ { "operation": "inbox_queue", "name": "Correspondence queue triage", "purpose": "Use when a correspondence officer asks what is waiting to be answered, soonest due first." }, { "operation": "break_down_enquiry", "name": "Enquiry breakdown", "purpose": "Use when an officer wants one letter split into its separate questions so none is missed." }, { "operation": "find_precedents", "name": "Approved precedent match", "purpose": "Use when an officer asks which approved past responses already answer each question." }, { "operation": "route_and_refer", "name": "Team routing and referral", "purpose": "Use when an officer asks who owns each part and what to ask the team about an unanswered part." }, { "operation": "draft_response", "name": "Cited draft reply", "purpose": "Use when an officer asks for a draft reply that answers every question with citations." }, { "operation": "library_update", "name": "Precedent library update", "purpose": "Use when someone asks how newly approved wording becomes reusable so the question is not re-asked." } ], "easy_mode": [ "Verify Power Platform CLI 2.9.3 or newer and authenticate to the approved target environment.", "Initialize a modern cli-copilot project with an approved publisher prefix and preserve the generated identity settings.", "Upload the two packaged synthetic knowledge files and every operation-specific SKILL.md.", "Bind only approved least-privilege production connections after security, privacy, and business-owner review.", "Push the Draft agent, replay every locked persona case, and stop before publish for explicit approval." ], "local_install_prompt": "Install Public Correspondence Agent from https://raw.githubusercontent.com/microsoft/aibast-agents-library/main/solutions/public-correspondence-response/deployment.json, verify the expected tool and smoke case, and preserve the synthetic read-only boundary. Do not ask me to open a terminal. Do not connect customer systems or execute external actions.", "copilot_studio_prompt": "Using the Microsoft Copilot Studio plugin, initialize a Draft agent for Public Correspondence Agent from https://raw.githubusercontent.com/microsoft/aibast-agents-library/main/solutions/public-correspondence-response/deployment.json. Upload the two synthetic knowledge files and 6 skills, replay every locked case, and Stop before publish. Do not bind production connections without explicit customer approval.", "required_connections": [ "Use only approved, least-privilege connections to correspondence records and the approved-response library; the packaged demonstration uses synthetic knowledge files." ], "manual_commands": [ "pac auth create", "pac copilot init --name \"Public Correspondence Agent\" --publisher-prefix --authoring-mode cli-copilot --project-dir \"\" --environment \"\"", "pac connection list --environment \"\"", "pac copilot pull --project-dir \"\"", "pac copilot push --project-dir \"\"", "pac copilot publish --bot \"\" --environment \"\"" ], "acceptance_checks": [ "Every exposed operation has one persona-language case and one SKILL.md.", "Exactly two synthetic knowledge files are present.", "Responses state the relevant human approval or review boundary.", "No operation claims an external write, communication, commitment, or protected decision.", "Publishing remains a separate user-approved step." ], "hard_mode": [ "Run pac copilot init and verify settings.mcs.yml before authoring components.", "Author global instructions that preserve the packaged synthetic-only and human-approval boundaries.", "Add the two knowledge files and one skill for every exposed operation.", "Create dedicated connection references only for approved systems and verify least-privilege access.", "Pull before push, inspect every generated file, replay every locked case, and stop before publish." ] }, "curated_copy": true, "source": "field", "origin": "both", "onepager": "58. Public Correspondence Agent one-pager.pptx", "demo_video": "58-Public Correspondence Agent.mp4", "has_onepager": true, "has_demo_video": true, "promise_coverage": 100, "is_primary": true, "aliases": [], "package": { "slug": "public-correspondence-response", "quest_url": "solutions/public-correspondence-response/quest.html", "manual_tutorial_url": "solutions/public-correspondence-response/manual-tutorial.html", "export_manifest_url": "solutions/public-correspondence-response/export-manifest.json" } }, "_demo": { "slug": "public-correspondence-response", "case_count": 6, "personas": [ "Correspondence Officer", "Correspondence Team Lead", "Knowledge Owner" ], "prompts": [ "What correspondence do I need to answer this week?", "Break down the scanned letter from E. Okafor into its separate questions.", "Which approved past responses already answer each question in the Okafor letter?" ] }, "_catalog_kind": "solution", "_readiness": [ "Demo proven", "Conversation tested", "Synthetic data", "Production pattern" ] }, { "schema": "rapp-agent/1.0", "name": "@aibast-agents-library/utility-billing-assistance", "version": "1.1.0", "display_name": "Utility Billing and Assistance Agent", "description": "Review synthetic municipal utility accounts for billing questions, smart-meter usage anomalies, draft payment-plan options, and preliminary assistance-program screening. The agent never adjusts a bill, enrolls a resident, starts a payment plan, schedules a repair, or changes an account; authorized utility staff must approve actions through future authenticated tools.", "author": "AIBAST", "tags": [ "utility", "billing", "water", "payment", "assistance", "municipal" ], "category": "slg_government", "quality_tier": "verified", "requires_env": [], "dependencies": [ "@rapp/basic-agent" ], "_file": "agents/@aibast-agents-library/slg_government_stacks/utility_billing_assistance_stack/utility_billing_assistance_agent.py", "_size_kb": 39.0, "_lines": 747, "_sha256": "ed02f2e42c47c1e04bf331e0941be5f59f9887fc9e268a520597d8819f33f33f", "_install_prefix": "utility_billing_assistance__", "_install_filename": "utility_billing_assistance__ed02f2e42c47_agent.py", "_stack": "utility_billing_assistance", "_stack_vertical": "slg_government", "_synthetic_data": true, "_added_at": "2026-03-18T13:23:12-04:00", "_solution": { "advertised_name": "Utility Billing and Assistance Agent", "sharepoint_list_name": "Utility Leak Detection & Billing Agent", "slot": 24, "executive_summary": "Automate leak detection and billing processes to improve customer satisfaction, ensure policy compliance, and protect municipal revenue.", "industries": [ "Energy and Utilities", "State and Local Government" ], "personas": [ "Customer Service Rep", "Billing Specialist", "Assistance Coordinator" ], "featured_tools": [ "Dynamics 365", "Microsoft Teams", "Outlook" ], "agent_requirements": [ "Microsoft Copilot Studio" ], "capabilities": [ "Automate smart meter analysis for leak detection and usage anomalies", "Apply municipal code rules to calculate credits and adjust bills", "Screen income eligibility and pre-fill applications for assistance programs", "Set up payment plans and schedule repairs in one efficient workflow" ], "outcomes": [ "Resolved complex billing disputes in a single interaction", "Increased assistance enrollment while reducing delinquent accounts", "Eliminated cross-system manual checks to improve productivity and consistency", "Strengthened customer trust with transparent, compliant resolution" ], "customer_scenario": [ "Manual review of consumption data delayed resolutions and led to frustrated residents", "Inconsistently applied leak adjustment policies resulted in added compliance risks", "Customers missed assistance opportunities due to slow, complex screening and enrollment processes" ], "slide_title": "Utility Billing and Assistance Agent", "slide_summary": "Automates leak detection and billing processes to improve customer satisfaction, ensure policy compliance, and protect municipal revenue.", "scenario_name": "A city utility provider", "challenge_intro": "A utility provider faced rising customer complaints and billing disputes:", "actions_intro": "The utility provider deployed the Utility Billing and Assistance Agent to:", "outcomes_intro": "With the agent\u2019s help, the utility provider:", "opportunity_statements": [ "Analyze smart meter data to identify anomalies and diagnose leaks for rapid dispute resolution", "Apply municipal policies automatically to calculate adjustments and create full audit trails", "Screen eligibility for assistance programs and configure payment plans to maximize financial relief" ], "module_name": null, "module_description": null, "source_verified": true, "sales_headline": "Resolve billing questions with evidence while keeping account authority human", "card_pitch": "Bring synthetic balances, smart-meter history, leak-policy screening, payment options, and assistance criteria into one resident-service brief without changing a bill, plan, repair, enrollment, or account.", "why_try": "A billing specialist can ask about a disputed account and see the exact usage pattern, draft estimate, required evidence, and approval boundary in one response.", "customer_challenge_copy": "Consumption records, billing policy, assistance criteria, and account status are reviewed manually across systems, slowing dispute resolution and creating inconsistent resident outcomes.", "microsoft_ai_story": "Microsoft Copilot Studio can coordinate governed review across a utility billing platform, advanced metering infrastructure, assistance case management, Outlook, Teams, and approved repair scheduling workflows.", "business_value": [ "Improves consistency of usage and billing-evidence review.", "Makes assistance and payment-plan criteria easier to explain.", "Preserves resident and municipal controls by keeping every account action behind authorized staff approval." ], "search_terms": [ "utility billing assistance", "leak adjustment", "smart meter anomaly", "billing dispute", "payment plan", "utility assistance eligibility", "municipal utility", "advanced metering infrastructure", "Dynamics 365", "Microsoft Teams", "Outlook", "Copilot Studio" ], "journey_stage": "Serve", "blueprint_role": "Creates a governed resident-service evidence layer before any billing adjustment, eligibility decision, enrollment, payment agreement, repair schedule, or account update.", "sample_prompts": [ { "label": "Customer Service Representative: Read-only account explanation", "prompt": "Explain ACCT-90003 balances without changing the account.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=utility-billing-assistance-billing-inquiry" }, { "label": "Billing Specialist: Leak and usage screening", "prompt": "Does ACCT-90003 show a possible leak and what draft adjustment evidence is needed?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=utility-billing-assistance-usage-analysis" }, { "label": "Revenue Services Supervisor: Draft payment-plan options", "prompt": "Show ACCT-90003 payment-plan options, but do not set one up.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=utility-billing-assistance-payment-plan" }, { "label": "Assistance Coordinator: Resident assistance eligibility", "prompt": "What assistance programs could help the Maple Drive resident on a fixed income, and do they qualify?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=utility-billing-assistance-assistance-programs" }, { "label": "Leak Adjustment Analyst: Leak adjustment credit draft", "prompt": "Does the Maple Drive household qualify for a leak credit, and how much would it be?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=utility-billing-assistance-leak-adjustment" }, { "label": "Enrollment Specialist: Payment plan and application packet", "prompt": "Get the payment plan and the water-assistance application ready for the Maple Drive resident. What paperwork do they need?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=utility-billing-assistance-application-packet" }, { "label": "Water Conservation Coordinator: Repair assistance program", "prompt": "Can we help the Maple Drive resident fix the leak at no cost?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=utility-billing-assistance-repair-assistance" }, { "label": "Customer Resolution Manager: Complete assistance package", "prompt": "Pull everything together for the Maple Drive resident so we can send it and close out the account notes.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=utility-billing-assistance-resolution-summary" }, { "label": "Eligibility Screener: Preliminary household screening", "prompt": "Screen a two-person household earning 25000 with a 70-year-old applicant.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=utility-billing-assistance-income-screen" } ], "architecture": { "business_flow": [ "Customer service representatives, billing specialists, assistance coordinators, and revenue supervisors", "Microsoft 365 Copilot or Copilot Studio", "Utility Billing and Assistance Agent", "Utility billing system", "Advanced metering infrastructure", "Assistance case management, Outlook, and Teams", "Authorized utility staff and resident" ], "capabilities": [ { "operation": "billing_inquiry", "name": "Read-only account explanation", "purpose": "Summarizes synthetic balances, services, payment history, and account state without changing a record." }, { "operation": "usage_analysis", "name": "Leak and usage screening", "purpose": "Reviews synthetic meter history and drafts a policy estimate without diagnosing a leak or adjusting a bill." }, { "operation": "payment_plan", "name": "Draft payment-plan options", "purpose": "Calculates non-binding installment options without creating an agreement." }, { "operation": "assistance_programs", "name": "Resident assistance eligibility", "purpose": "Screens the resident's income on file against synthetic program criteria and totals potential relief without determining eligibility or enrolling anyone." }, { "operation": "leak_adjustment", "name": "Leak adjustment credit draft", "purpose": "Applies the Municipal Code 18.42 one-time leak adjustment to the smart-meter excess and drafts the credit and new bill for billing-specialist approval without adjusting the bill." }, { "operation": "application_packet", "name": "Payment plan and application packet", "purpose": "Drafts the payment plan and the pre-filled assistance application with required documents without creating a plan or submitting an application." }, { "operation": "repair_assistance", "name": "Repair assistance program", "purpose": "Matches the resident to the free repair program and proposes an appointment without booking it." }, { "operation": "resolution_summary", "name": "Complete assistance package", "purpose": "Assembles the credit, plan, applications, and repair into a customer-ready draft and account notes without sending or changing anything." }, { "operation": "income_screen", "name": "Preliminary household screening", "purpose": "Compares supplied household facts with synthetic criteria without determining eligibility or enrolling an applicant." } ], "easy_mode": [ "Verify Power Platform CLI 2.9.3 or newer and authenticate to the target environment.", "Use the Copilot Studio plugin initialization workflow to create a modern, sync-connected Draft agent with an approved publisher prefix.", "Upload manual/GLOBAL-INSTRUCTIONS.md, the two synthetic knowledge files, and all 4 operation skills from the solution package.", "Bind only approved, least-privilege production connections through dedicated connection references and preserve the human approval boundary.", "Push the Draft project, replay every locked persona prompt, inspect safety behavior, and stop before publishing." ], "local_install_prompt": "You are GitHub Copilot running in Agent mode in VS Code. Install and open the AI BAST Utility Billing and Assistance Agent from this deployment recipe: https://raw.githubusercontent.com/microsoft/aibast-agents-library/main/solutions/utility-billing-assistance/deployment.json. Own the setup and do not ask me to open a terminal. Verify the Brainstem health endpoint, install or start it if needed using the official recipe, fetch the registry and source, verify the source hash, place the exact agent file in the discovered agents directory, confirm the expected tool, run the smoke prompt, and open the local UI. Preserve the synthetic read-only boundary. Do not connect customer systems, perform an external action, publish, or silently substitute another agent. Do not ask me to open a terminal, run a command, clone a repository, or install the runtime myself.", "copilot_studio_prompt": "You are GitHub Copilot running in Agent mode in VS Code with the Microsoft Copilot Studio plugin installed. Promote the AI BAST Utility Billing and Assistance Agent using this deployment recipe: https://raw.githubusercontent.com/microsoft/aibast-agents-library/main/solutions/utility-billing-assistance/deployment.json. Ask only for the target environment ID and publisher prefix. Initialize a modern cli-copilot Draft project, upload the packaged global instructions, two synthetic knowledge files, and 4 skills, and preserve these operations: billing_inquiry, usage_analysis, payment_plan, assistance_programs. Bind only approved dedicated connections, review every generated file and connection, pull before push, replay every locked persona case, enforce explicit authorized approval for any future write, and stop before publish. Show validation evidence and unresolved connection work. Stop before publish.", "required_connections": [ "Utility billing and customer information system", "Advanced metering infrastructure", "Assistance case-management system", "Approved payment-plan and repair workflows", "Microsoft Outlook and Teams" ], "manual_commands": [ "pac auth create", "pac copilot init --name \"Utility Billing and Assistance Agent\" --publisher-prefix --authoring-mode cli-copilot --project-dir \"\" --environment \"\"", "pac connection list --environment \"\"", "pac copilot pull --project-dir \"\"", "pac copilot push --project-dir \"\"", "pac copilot publish --bot \"\" --environment \"\"" ], "acceptance_checks": [ "settings.mcs.yml preserves the initialized identity, environment, language, and authoring mode.", "All 4 implemented operations are represented by instructions, a packaged skill, or a bound tool.", "Exactly two synthetic knowledge files are loaded and clearly labeled fictional.", "Every production source uses a dedicated, least-privilege connection reference.", "Every locked persona prompt returns its deterministic synthetic evidence and the relevant human approval boundary.", "Never adjust a bill, determine final eligibility, enroll a resident, or start a payment plan.", "Never schedule a repair or change an account.", "Public value statements remain qualitative; exact figures appear only as synthetic evidence.", "Publishing remains a separate, user-approved step after push and validation." ], "hard_mode": [ "Run pac copilot init and confirm settings.mcs.yml exists.", "Author global instructions that route every operation and preserve the packaged no-write boundaries.", "Upload the two synthetic Markdown knowledge files and 4 operation-specific SKILL.md files.", "Create read tools first; isolate any future write action behind authorization, current-state validation, explicit confirmation, and immutable audit logging.", "Validate every component schema and filename, push the Draft, replay every locked prompt, inspect failures, and stop before publish." ] }, "curated_copy": true, "source": "catalog", "origin": null, "onepager": "27. Utility Billing and Assistance Agent one-pager.pptx", "demo_video": "27-Utility Billing and Assistance Agent.mp4", "has_onepager": true, "has_demo_video": true, "promise_coverage": 30, "is_primary": true, "aliases": [], "package": { "slug": "utility-billing-assistance", "quest_url": "solutions/utility-billing-assistance/quest.html", "manual_tutorial_url": "solutions/utility-billing-assistance/manual-tutorial.html", "export_manifest_url": "solutions/utility-billing-assistance/export-manifest.json" } }, "_demo": { "slug": "utility-billing-assistance", "case_count": 9, "personas": [ "Assistance Coordinator", "Billing Specialist", "Customer Resolution Manager", "Customer Service Representative", "Eligibility Screener", "Enrollment Specialist", "Leak Adjustment Analyst", "Revenue Services Supervisor", "Water Conservation Coordinator" ], "prompts": [ "Explain ACCT-90003 balances without changing the account.", "Does ACCT-90003 show a possible leak and what draft adjustment evidence is needed?", "Show ACCT-90003 payment-plan options, but do not set one up." ] }, "_catalog_kind": "solution", "_readiness": [ "Demo proven", "Conversation tested", "Synthetic data", "Production pattern" ] }, { "schema": "rapp-agent/1.0", "name": "@aibast-agents-library/license-renewal-expansion", "version": "1.0.0", "display_name": "License Renewal and Expansion Agent", "description": "Streamline subscription renewal management and expansion planning, turning risk into growth opportunities while increasing win probability.", "author": "AIBAST", "tags": [ "license", "renewal", "expansion", "churn", "revenue", "saas" ], "category": "software_digital_products", "quality_tier": "verified", "requires_env": [], "dependencies": [ "@rapp/basic-agent" ], "_file": "agents/@aibast-agents-library/software_dp_stacks/license_renewal_expansion_stack/license_renewal_expansion_agent.py", "_size_kb": 36.1, "_lines": 800, "_sha256": "fd97d48a5ec515d3308e1d0358aa822a3559a12609832fd072992666c59c80d6", "_install_prefix": "license_renewal_expansion__", "_install_filename": "license_renewal_expansion__fd97d48a5ec5_agent.py", "_stack": "license_renewal_expansion", "_stack_vertical": "software_digital_products", "_synthetic_data": true, "_added_at": "2026-03-18T13:23:12-04:00", "_solution": { "advertised_name": "License Renewal and Expansion Agent", "sharepoint_list_name": "Subscription Renewal Agent", "slot": 33, "executive_summary": "Streamline subscription renewal management and expansion planning, turning risk into growth opportunities while increasing win probability.", "industries": [ "Software Tech" ], "personas": [ "Account Executives", "Customer Success Managers", "Sales Leadership" ], "featured_tools": [ "Dynamics 365", "Microsoft Teams" ], "agent_requirements": [ "Microsoft Copilot Studio" ], "capabilities": [ "Analyze customer health, feature adoption, usage spikes, and waitlisted demand", "Compare competitive offers and calculate true switching costs", "Identify credible expansion opportunities tied to user demand", "Get a pricing and packaging strategy with pre\u2011approved concessions" ], "outcomes": [ "Delivered a tailored renewal and expansion proposal", "Countered competitive threats with data\u2011driven switching\u2011cost analysis", "Secured expansion by validating unmet demand and aligning pricing to requirements", "Accelerated internal approvals and positioned the renewal as a multi\u2011year strategic partnership" ], "customer_scenario": [ "The account showed strong usage but was targeted by a competitor", "Procurement requested competing bids, raising pricing pressure", "Manual proposal creation slowed response times", "The sales team needed stronger justification for value delivered and a strategic pitch for expansion" ], "slide_title": "License Renewal and Expansion Agent", "slide_summary": "Streamlines subscription renewal management and expansion planning, turning risk into growth opportunities while increasing win probability.", "scenario_name": "An enterprise SaaS provider", "challenge_intro": "An enterprise SaaS provider faced a high\u2011stakes renewal situation:", "actions_intro": "The SaaS provider used the License Renewal and Expansion Agent to:", "outcomes_intro": "With the agent, the company:", "opportunity_statements": [ "Streamlines subscription renewal management and expansion planning, turning risk into growth opportunities while increasing win probability.", "Assess account health and usage signals for targeted, proactive renewal planning", "Identify competitive threats and switching costs to support negotiation strategy", "Generate optimized renewal pricing and packaging so teams can close deals faster", "Draft communications and renewal materials for streamlined preparation and engagement" ], "module_name": null, "module_description": null, "source_verified": true, "sales_headline": "Bring renewal risk and expansion evidence into one review", "card_pitch": "Help account, customer-success, and sales leaders review renewal timing, account health, demand signals, churn evidence, and clearly labeled commercial scenarios before negotiation begins.", "why_try": "A team can inspect the evidence behind a renewal or expansion recommendation while keeping pricing, concessions, contracts, and customer engagement under authorized control.", "customer_challenge_copy": "Renewal planning is often fragmented across usage, support, health, demand, competition, and commercial assumptions, slowing coordinated decisions.", "microsoft_ai_story": "Microsoft Copilot Studio provides the renewal-planning workspace. Dynamics 365 Sales contributes account and renewal context, approved product-telemetry and entitlement sources contribute usage evidence, and Microsoft Teams supports customer-success and commercial review.", "business_value": [ "Creates a shared renewal and risk review for sales and customer-success teams.", "Makes demand and churn signals easier to connect to planning options.", "Supports commercial scenario review without automating negotiation or outreach." ], "search_terms": [ "license renewal", "subscription renewal", "expansion planning", "churn risk", "usage signals", "switching costs", "renewal pricing", "Dynamics 365 Sales", "product telemetry", "customer success", "Microsoft Teams sales", "Copilot Studio" ], "journey_stage": "Retain", "blueprint_role": "Provides a governed planning layer between subscription evidence and authorized renewal or expansion execution.", "sample_prompts": [ { "label": "Review the renewal pipeline", "prompt": "Review the bundled synthetic renewal pipeline, risk bands, and preparation checklist without changing CRM or forecast records.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=license-renewal-expansion-renewal-pipeline" }, { "label": "Find expansion signals", "prompt": "Identify synthetic demand signals and draft packaging options that still require authorized pricing review.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=license-renewal-expansion-expansion-opportunities" }, { "label": "Assess churn and switching risk", "prompt": "Which synthetic accounts show churn or competitor risk, and what switching-cost assumptions require validation?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=license-renewal-expansion-churn-risk" }, { "label": "Compare renewal scenarios", "prompt": "Compare the bundled synthetic renewal, expansion, and churn scenarios without making revenue commitments.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=license-renewal-expansion-revenue-impact" }, { "label": "Account Executive: Strategic account analysis", "prompt": "GlobalBank license expires in 45 days. Currently 2,000 seats at $1M ARR. Usage shows they need 500 more seats. Competitor offering 30% discount.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=license-renewal-expansion-account-health" }, { "label": "Customer Success Manager: Competitive defense and switching costs", "prompt": "Yes, show me the competitive defense strategy.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=license-renewal-expansion-competitive-defense" }, { "label": "Account Executive: Renewal and expansion proposal draft", "prompt": "Yes, build the renewal and expansion proposal.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=license-renewal-expansion-renewal-proposal" }, { "label": "Account Executive: Executive presentation outline", "prompt": "Yes, create the executive presentation with talking points.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=license-renewal-expansion-executive-brief" }, { "label": "Sales Leadership: Negotiation boundaries and approvals", "prompt": "Yes, show me negotiation strategy and what approvals I need.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=license-renewal-expansion-negotiation-plan" }, { "label": "Sales Leadership: Renewal strategy summary", "prompt": "Yes, send the proposal and summarize our renewal strategy.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=license-renewal-expansion-deal-summary" } ], "architecture": { "business_flow": [ "Account executive, customer success manager, or sales leader", "Copilot experience", "License Renewal and Expansion Agent", "Synthetic subscription and usage evidence", "Human renewal and commercial review" ], "capabilities": [ { "operation": "renewal_pipeline", "name": "Renewal Pipeline", "purpose": "Organizes renewal timing, health, ownership, and risk evidence." }, { "operation": "expansion_opportunities", "name": "Expansion Opportunities", "purpose": "Surfaces demand signals and draft expansion options." }, { "operation": "churn_risk", "name": "Churn Risk", "purpose": "Explains usage, support, stakeholder, and competition risk signals." }, { "operation": "revenue_impact", "name": "Revenue Impact", "purpose": "Compares labeled renewal, expansion, and churn scenarios." }, { "operation": "account_health", "name": "Strategic account analysis", "purpose": "Shows one account's ARR, seat usage, waitlist demand, health, competitor threat and value realized." }, { "operation": "competitive_defense", "name": "Competitive defense and switching costs", "purpose": "Compares a competitor offer and itemizes the customer's true switching costs with a draft counter-strategy." }, { "operation": "renewal_proposal", "name": "Renewal and expansion proposal draft", "purpose": "Prices the renewal and waitlist expansion with term options and customer ROI as a draft for pricing review." }, { "operation": "executive_brief", "name": "Executive presentation outline", "purpose": "Outlines the executive presentation, talking points and objection handlers as a draft." }, { "operation": "negotiation_plan", "name": "Negotiation boundaries and approvals", "purpose": "Lists concession boundaries, required approvers and negotiation strategy without granting approvals." }, { "operation": "deal_summary", "name": "Renewal strategy summary", "purpose": "Summarizes the renewal strategy and deal metrics with the proposal ready for the user to send." } ], "easy_mode": [ "Verify Power Platform CLI and authenticate to the target environment.", "Initialize a modern cli-copilot project with an approved publisher prefix.", "Upload the two synthetic knowledge files and create one manual skill for every advertised operation.", "Bind dedicated approved connection references only after replacing the synthetic evidence route.", "Push the draft, replay every locked persona case, and stop before publication or any external action." ], "local_install_prompt": "You are GitHub Copilot running in Agent mode in VS Code. Install and open the AI BAST License Renewal and Expansion Agent from this deployment recipe: https://raw.githubusercontent.com/microsoft/aibast-agents-library/main/solutions/license-renewal-expansion/deployment.json. Own the setup and do not ask me to run terminal commands. Inspect the operating system; verify the Brainstem health endpoint; if needed, use the official installer from the recipe; fetch and install the portable source; verify the expected tool; and replay the recipe smoke test. Preserve the synthetic-only route for renewal_pipeline, expansion_opportunities, churn_risk, revenue_impact. Do not connect customer systems, perform writes, publish, approve pricing, send outreach, or silently substitute another agent. If authentication requires browser confirmation, ask only for that browser step. Stop and report the exact evidence and any failed step. Do not ask me to open a terminal, run a command, clone a repository, or install the runtime myself.", "copilot_studio_prompt": "You are GitHub Copilot running in Agent mode in VS Code with the Microsoft Copilot Studio plugin installed. Prepare the AI BAST License Renewal and Expansion Agent from https://raw.githubusercontent.com/microsoft/aibast-agents-library/main/solutions/license-renewal-expansion/deployment.json. Ask only for the target Power Platform environment ID and publisher prefix. Verify PAC CLI, authenticate if needed, initialize a cli-copilot project, upload the two synthetic knowledge files, and create one skill for each operation: renewal_pipeline, expansion_opportunities, churn_risk, revenue_impact. Bind only dedicated approved connection references, preserve every evidence and approval boundary, pull before push, and replay every locked persona case. Stop before publish and show the generated files, validation evidence, unresolved connection work, and the exact publish command that would require user approval.", "required_connections": [ "Dynamics 365 Sales", "Subscription and entitlement system", "Approved product-telemetry source", "Customer-success platform", "Authorized pricing and concession workflow", "Microsoft Teams" ], "manual_commands": [ "pac auth create", "pac copilot init --name \"License Renewal and Expansion Agent\" --publisher-prefix --authoring-mode cli-copilot --project-dir \"\" --environment \"\"", "pac connection list --environment \"\"", "pac copilot pull --project-dir \"\"", "pac copilot push --project-dir \"\"", "pac copilot publish --bot \"\" --environment \"\"" ], "acceptance_checks": [ "The package remains under solutions/license-renewal-expansion and the portable source resolves to the expected tool.", "Every advertised operation has one uploadable skill and one locked persona-language capture case.", "The approved one-pager filename and source hash remain pinned in the one-pager map.", "Every exact value is identified as synthetic or as a planning assumption rather than a measured customer outcome.", "No CRM write, outreach, approval, pricing change, forecast change, task, alert, document delivery, or customer communication occurs.", "Publishing remains a separate user-approved step after push and validation." ], "hard_mode": [ "Run pac copilot init and confirm settings.mcs.yml preserves the intended environment, language, and publisher identity.", "Author global instructions that route only the supported operations: renewal_pipeline, expansion_opportunities, churn_risk, revenue_impact.", "Upload both synthetic knowledge files and create each operation skill from the package SKILL.md source.", "Create dedicated connection references for the approved production seams; do not reuse ambiguous existing connections.", "Validate schema, instructions, skills, knowledge, and connection bindings; pull before push and replay every locked case.", "Keep the agent in Draft and require explicit confirmation before publish or any external action." ] }, "curated_copy": true, "source": "catalog", "origin": null, "onepager": "36. License Renewal and Expansion Agent one-pager.pptx", "demo_video": "36-License Renewal and Expansion Agent.mp4", "has_onepager": true, "has_demo_video": true, "promise_coverage": 62, "is_primary": true, "aliases": [], "package": { "slug": "license-renewal-expansion", "quest_url": "solutions/license-renewal-expansion/quest.html", "manual_tutorial_url": "solutions/license-renewal-expansion/manual-tutorial.html", "export_manifest_url": "solutions/license-renewal-expansion/export-manifest.json" } }, "_demo": { "slug": "license-renewal-expansion", "case_count": 10, "personas": [ "Account Executive", "Customer Success Manager", "Sales Leadership" ], "prompts": [ "Review the bundled synthetic renewal pipeline, risk bands, and preparation checklist without changing CRM or forecast records.", "Identify synthetic demand signals and draft packaging options that still require authorized pricing review.", "Which synthetic accounts show churn or competitor risk, and what switching-cost assumptions require validation?" ] }, "_catalog_kind": "solution", "_readiness": [ "Demo proven", "Conversation tested", "Synthetic data", "Production pattern" ] }, { "schema": "rapp-agent/1.0", "name": "@aibast-agents-library/product-feedback-synthesizer", "version": "1.0.0", "display_name": "Product Feedback Synthesizer Agent", "description": "Turn fragmented feedback into actionable insights that accelerate product improvements, prevent churn, and optimize engineering priorities.", "author": "AIBAST", "tags": [ "feedback", "product", "feature-requests", "sentiment", "roadmap", "nps" ], "category": "software_digital_products", "quality_tier": "verified", "requires_env": [], "dependencies": [ "@rapp/basic-agent" ], "_file": "agents/@aibast-agents-library/software_dp_stacks/product_feedback_synthesizer_stack/product_feedback_synthesizer_agent.py", "_size_kb": 22.4, "_lines": 429, "_sha256": "232a8e924f6dc9869d40990a6a225a8509b626a8a23a938120f4b7201cb6269d", "_install_prefix": "product_feedback_synthesizer__", "_install_filename": "product_feedback_synthesizer__232a8e924f6d_agent.py", "_stack": "product_feedback_synthesizer", "_stack_vertical": "software_digital_products", "_synthetic_data": true, "_added_at": "2026-03-18T13:23:12-04:00", "_solution": { "advertised_name": "Product Feedback Synthesizer Agent", "sharepoint_list_name": "Product Feedback Insights Agent", "slot": 23, "executive_summary": "Turn fragmented feedback into actionable insights that accelerate product improvements, prevent churn, and optimize engineering priorities.", "industries": [ "Cross-Industry" ], "personas": [ "Product Manager", "Engineering Lead", "Director of Product" ], "featured_tools": [ "Dynamics 365", "Microsoft Teams" ], "agent_requirements": [ "Microsoft Copilot Studio" ], "capabilities": [ "Aggregate customer feedback from diverse sources into a single view", "Analyze sentiment and identify top pain points and feature requests", "Prioritize items based on impact and automate Jira ticket creation", "Alert teams to churn risks and competitive gaps for proactive intervention" ], "outcomes": [ "Reduced feedback analysis time from weeks to minutes", "Delivered data-driven prioritization for roadmap planning", "Prevented churn through early detection of at-risk accounts", "Accelerated feature delivery and improved customer satisfaction" ], "customer_scenario": [ "Manual analysis delayed roadmap decisions and missed critical issues", "Subjective prioritization led to misaligned engineering investments", "Teams lacked visibility into churn risks until cancellations occurred", "Competitive gaps blocked deals without timely product response" ], "slide_title": "Product Feedback Synthesizer Agent", "slide_summary": "Turns fragmented feedback into actionable insights that accelerate product improvements, prevent churn, and optimize engineering priorities.", "scenario_name": "An enterprise SaaS company", "challenge_intro": "A SaaS company struggled to manage overwhelming feedback volume across multiple channels:", "actions_intro": "The company leveraged the Product Feedback Synthesizer Agent to:", "outcomes_intro": "With the agent\u2019s help, the company:", "opportunity_statements": [ "Aggregate and analyze feedback from multiple sources to provide a unified view of customer sentiment and priorities", "Apply AI-driven sentiment and pattern recognition to identify critical pain points and feature requests", "Automate prioritization and workflow orchestration to convert insights into engineering actions and roadmap updates" ], "module_name": null, "module_description": null, "source_verified": true, "sales_headline": "Turn feedback noise into evidence the product trio can debate", "card_pitch": "Unify fictional feedback, sentiment, feature requests, and impact signals into transparent review candidates without turning one score or account comment into a roadmap promise.", "why_try": "The demo lets a product manager ask a natural question and receive a source-backed, reviewable answer with the approval boundary visible in the response.", "customer_challenge_copy": "Product teams spend too long reconciling channels, while subjective prioritization and isolated anecdotes can obscure broader patterns and tradeoffs.", "microsoft_ai_story": "Copilot Studio can synthesize governed signals from Dynamics 365, support, survey, and collaboration sources, with Teams bringing product, engineering, design, and commercial owners into review.", "business_value": [ "Reduces manual effort in cross-channel synthesis.", "Makes prioritization inputs and assumptions easier to inspect.", "Supports earlier, cross-functional review of product signals." ], "search_terms": [ "Director of Product", "Engineering Lead", "Microsoft Copilot Studio", "Microsoft Teams", "Product Feedback Synthesizer Agent", "Product Manager", "feature_requests", "feedback_summary", "roadmap_impact", "sentiment_analysis" ], "journey_stage": "Innovate", "blueprint_role": "Creates a governed evidence layer between fragmented feedback systems and human product strategy.", "sample_prompts": [ { "label": "Product Manager: Cross-channel feedback summary", "prompt": "Give me the cross-channel feedback picture before the weekly product review.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=product-feedback-synthesizer-feedback-summary" }, { "label": "Engineering Lead: Feature-request evidence ranking", "prompt": "Rank the feature-request evidence for review, but do not turn it into a roadmap.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=product-feedback-synthesizer-feature-requests" }, { "label": "Director of Product: Sentiment and NPS evidence", "prompt": "What are the strongest sentiment signals in the fictional customer snapshot?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=product-feedback-synthesizer-sentiment-analysis" }, { "label": "Product Manager: Roadmap review candidates", "prompt": "Frame the impact tradeoffs the product trio should validate before sequencing work.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=product-feedback-synthesizer-roadmap-impact" } ], "architecture": { "business_flow": [ "Product, engineering, design, and customer teams", "Microsoft 365 Copilot or Copilot Studio", "Product Feedback Synthesizer Agent", "CRM, support, survey, and product feedback sources", "Microsoft Teams product review", "Human roadmap governance" ], "capabilities": [ { "operation": "feedback_summary", "name": "Cross-channel feedback summary", "purpose": "Use when a product manager asks for a compact view of volume, channels, categories, and synthetic commercial context." }, { "operation": "feature_requests", "name": "Feature-request evidence ranking", "purpose": "Use when an engineering lead wants review candidates sorted by the supplied synthetic ARR weight." }, { "operation": "sentiment_analysis", "name": "Sentiment and NPS evidence", "purpose": "Use when a director of product asks for the fictional sentiment split and recent excerpts." }, { "operation": "roadmap_impact", "name": "Roadmap review candidates", "purpose": "Use when a product trio wants a non-binding impact view with explicit validation gates." } ], "easy_mode": [ "Verify Power Platform CLI 2.9.3 or newer and authenticate to the approved target environment.", "Initialize a modern cli-copilot project with an approved publisher prefix and preserve the generated identity settings.", "Upload the two packaged synthetic knowledge files and every operation-specific SKILL.md.", "Bind only approved least-privilege production connections after security, privacy, and business-owner review.", "Push the Draft agent, replay every locked persona case, and stop before publish for explicit approval." ], "local_install_prompt": "Install Product Feedback Synthesizer Agent from https://raw.githubusercontent.com/microsoft/aibast-agents-library/main/solutions/product-feedback-synthesizer/deployment.json, verify the expected tool and smoke case, and preserve the synthetic read-only boundary. Do not ask me to open a terminal. Do not connect customer systems or execute external actions.", "copilot_studio_prompt": "Using the Microsoft Copilot Studio plugin, initialize a Draft agent for Product Feedback Synthesizer Agent from https://raw.githubusercontent.com/microsoft/aibast-agents-library/main/solutions/product-feedback-synthesizer/deployment.json. Upload the two synthetic knowledge files and 4 skills, replay every locked case, and Stop before publish. Do not bind production connections without explicit customer approval.", "required_connections": [ "Use only approved, least-privilege production connections; the packaged demonstration uses synthetic knowledge files." ], "manual_commands": [ "pac auth create", "pac copilot init --name \"Product Feedback Synthesizer Agent\" --publisher-prefix --authoring-mode cli-copilot --project-dir \"\" --environment \"\"", "pac connection list --environment \"\"", "pac copilot pull --project-dir \"\"", "pac copilot push --project-dir \"\"", "pac copilot publish --bot \"\" --environment \"\"" ], "acceptance_checks": [ "Every exposed operation has one persona-language case and one SKILL.md.", "Exactly two synthetic knowledge files are present.", "Responses state the relevant human approval or review boundary.", "No operation claims an external write, communication, commitment, or protected decision.", "Publishing remains a separate user-approved step." ], "hard_mode": [ "Run pac copilot init and verify settings.mcs.yml before authoring components.", "Author global instructions that preserve the packaged synthetic-only and human-approval boundaries.", "Add the two knowledge files and one skill for every exposed operation.", "Create dedicated connection references only for approved systems and verify least-privilege access.", "Pull before push, inspect every generated file, replay every locked case, and stop before publish." ] }, "curated_copy": true, "source": "catalog", "origin": null, "onepager": "26. Product Feedback Synthesizer Agent one-pager.pptx", "demo_video": "26-Product Feedback Synthesizer Agent.mp4", "has_onepager": true, "has_demo_video": true, "promise_coverage": 70, "is_primary": true, "aliases": [], "package": { "slug": "product-feedback-synthesizer", "quest_url": "solutions/product-feedback-synthesizer/quest.html", "manual_tutorial_url": "solutions/product-feedback-synthesizer/manual-tutorial.html", "export_manifest_url": "solutions/product-feedback-synthesizer/export-manifest.json" } }, "_demo": { "slug": "product-feedback-synthesizer", "case_count": 6, "personas": [ "Director of Product", "Engineering Lead", "Product Manager" ], "prompts": [ "Give me the cross-channel feedback picture before the weekly product review.", "Rank the feature-request evidence for review, but do not turn it into a roadmap.", "What are the strongest sentiment signals in the fictional customer snapshot?" ] }, "_catalog_kind": "solution", "_readiness": [ "Demo proven", "Conversation tested", "Synthetic data", "Production pattern" ] }, { "schema": "rapp-agent/1.0", "name": "@aibast-agents-library/campaign-delivery-optimizer", "version": "1.0.0", "display_name": "Campaign Delivery Optimizer Agent", "description": "Help ad sales and ad operations teams keep advertiser campaigns on plan from a synthetic weekly delivery snapshot: each campaign's delivered reach against plan with the gap explained, the placements with extra reach available this week, and the drafted recommendation for the sales lead to approve. The agent never books inventory, changes a campaign or placement, or shares anything with an advertiser; the sales lead approves every recommendation through future authenticated tools.", "author": "AIBAST", "tags": [ "media", "advertising", "ad-sales", "ad-operations", "campaign-delivery" ], "category": "telecom_media_gaming", "quality_tier": "verified", "requires_env": [], "dependencies": [ "@rapp/basic-agent" ], "_file": "agents/@aibast-agents-library/telecom_media_gaming_stacks/campaign_delivery_optimizer_stack/campaign_delivery_optimizer_agent.py", "_size_kb": 11.4, "_lines": 232, "_sha256": "817dea94625d6a15dda95fc418d3041f058c3ae12de0d406af7fa50878956c67", "_install_prefix": "campaign_delivery_optimizer__", "_install_filename": "campaign_delivery_optimizer__817dea94625d_agent.py", "_stack": "campaign_delivery_optimizer", "_stack_vertical": "telecom_media_gaming", "_synthetic_data": true, "_added_at": "2026-10-23T10:00:37-04:00", "_solution": { "advertised_name": "Campaign Delivery Optimizer Agent", "sharepoint_list_name": "Campaign Delivery Optimizer Agent", "slot": 69, "executive_summary": "Help ad sales and ad operations teams keep every advertiser campaign on plan by tracking weekly delivered reach against plan, finding placements with extra reach to close a delivery gap, and drafting the recommendation for the sales lead to approve, while nothing is booked or shared with an advertiser without a person's approval.", "industries": [ "Media & Entertainment", "Advertising" ], "personas": [ "Ad Operations Manager", "Campaign Manager", "Ad Sales Lead" ], "featured_tools": [ "Dynamics 365 Sales", "Microsoft Excel", "Microsoft Outlook" ], "agent_requirements": [ "Microsoft Copilot Studio" ], "capabilities": [ "Track each campaign's weekly reach against its plan", "Flag at-risk campaigns as soon as delivery falls behind", "Find placements with extra reach to close the gap", "Draft recommendations for the sales lead to approve and share" ], "outcomes": [ "Caught under-delivering campaigns within the week", "Recovered reach with the right placements", "Helped every campaign hit its goals", "Kept advertisers coming back" ], "customer_scenario": [ "Under-delivery surfaced late, after advertisers had noticed", "Teams pieced together delivery data from spreadsheets by hand", "Inventory that could recover reach was hard to find quickly", "Make-good proposals took days to draft and agree" ], "slide_title": "Campaign Delivery Optimizer Agent", "slide_summary": "Tracks every advertiser campaign against plan, finds extra reach where delivery falls short, and drafts the recommendation to share.", "scenario_name": "A national media company", "challenge_intro": "A national media company struggled to keep campaigns on plan:", "actions_intro": "The media company leveraged the Campaign Delivery Optimizer Agent to:", "outcomes_intro": "With the agent\u2019s help, the media company:", "opportunity_statements": [ "Assess campaign risk against plan each week to spot under-delivery before the advertiser does", "Find available reach across placements, such as a home-screen takeover, to close the gap", "Draft the recommendation for approval so the sales lead can review it and share it with the advertiser" ], "module_name": null, "module_description": null, "source_verified": true, "sales_headline": "Catch under-delivery within the week without handing advertiser commitments to a machine", "card_pitch": "Give ad sales and ad operations teams one conversational view of each campaign's delivery against plan, the placements with extra reach, and the drafted recommendation, while every recommendation stays with the sales lead to approve.", "why_try": "The demo lets an ad operations manager ask a natural question about a campaign and receive its delivery gap, the placements that can close it, and a draft recommendation, with the approval boundary visible in the response.", "customer_challenge_copy": "Media ad sales teams find under-delivery late, piece delivery data together by hand, and take days to draft make-good proposals.", "microsoft_ai_story": "Copilot Studio and Microsoft Teams give ad sales and ad operations a conversational assistant; Dynamics 365 Sales campaigns, Excel delivery reports, and Outlook drafts can be bound through least-privilege Power Platform connections with audit controls.", "business_value": [ "Flags at-risk campaigns as soon as weekly delivery falls behind plan.", "Recovers reach from available placements before the week closes.", "Cuts make-good drafting from days to a reviewable draft." ], "search_terms": [ "Ad Operations Manager", "Ad Sales Lead", "Campaign Delivery Optimizer Agent", "Campaign Manager", "Dynamics 365 Sales", "Microsoft Copilot Studio", "Microsoft Excel", "Microsoft Outlook", "ad inventory", "ad sales", "approval_recommendation", "available_reach", "campaign delivery", "campaign_risk", "make-good", "reach gap", "under-delivery" ], "journey_stage": "Optimize", "blueprint_role": "Separates delivery analysis and recommendation drafting from protected inventory booking, campaign changes, and advertiser communication.", "sample_prompts": [ { "label": "Ad Operations Manager: Campaign risk check", "prompt": "Is the Harborlight Streaming campaign falling behind this week, and by how much?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=campaign-delivery-optimizer-campaign-risk" }, { "label": "Campaign Manager: Available reach", "prompt": "Where can we find extra reach this week to close the Harborlight gap?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=campaign-delivery-optimizer-available-reach" }, { "label": "Ad Sales Lead: Recommendation draft for approval", "prompt": "Draft the make-good for Harborlight so Maya can review it. Don't send anything to the advertiser yet.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=campaign-delivery-optimizer-approval-recommendation" } ], "architecture": { "business_flow": [ "Ad sales leads, ad operations managers, and campaign managers", "Microsoft Teams or Microsoft 365 Copilot", "Campaign Delivery Optimizer Agent", "Dynamics 365 Sales campaigns and Excel delivery reports", "Available inventory placements", "Authorized sales-lead approval workflow" ], "capabilities": [ { "operation": "campaign_risk", "name": "Campaign risk check", "purpose": "Use when someone asks whether a campaign is behind plan this week, how big the delivery gap is, or why it is at risk." }, { "operation": "available_reach", "name": "Available reach", "purpose": "Use when someone asks which placements have extra reach this week to close a campaign's delivery gap." }, { "operation": "approval_recommendation", "name": "Recommendation draft for approval", "purpose": "Use when someone asks for the recommendation or make-good for a campaign, its approval status, or who approves it; returns a draft that is never shared." } ], "easy_mode": [ "Verify Power Platform CLI 2.9.3 or newer and authenticate to the approved target environment.", "Initialize a modern cli-copilot project with an approved publisher prefix and preserve the generated identity settings.", "Upload the two packaged synthetic knowledge files and every operation-specific SKILL.md.", "Bind only approved least-privilege production connections after security, privacy, and business-owner review.", "Push the Draft agent, replay every locked persona case, and stop before publish for explicit approval." ], "local_install_prompt": "Install Campaign Delivery Optimizer Agent from https://raw.githubusercontent.com/microsoft/aibast-agents-library/main/solutions/campaign-delivery-optimizer/deployment.json, verify the expected tool and smoke case, and preserve the synthetic read-only boundary. Do not ask me to open a terminal. Do not connect customer systems or execute external actions.", "copilot_studio_prompt": "Using the Microsoft Copilot Studio plugin, initialize a Draft agent for Campaign Delivery Optimizer Agent from https://raw.githubusercontent.com/microsoft/aibast-agents-library/main/solutions/campaign-delivery-optimizer/deployment.json. Upload the two synthetic knowledge files and 3 skills, replay every locked case, and Stop before publish. Do not bind production connections without explicit customer approval.", "required_connections": [ "Use only approved, least-privilege production connections; the packaged demonstration uses synthetic knowledge files.", "Dynamics 365 Sales campaigns and sales leads (read-only)", "Weekly delivery reports in Microsoft Excel (read-only)", "Available inventory and placement records (read-only)" ], "manual_commands": [ "pac auth create", "pac copilot init --name \"Campaign Delivery Optimizer Agent\" --publisher-prefix --authoring-mode cli-copilot --project-dir \"\" --environment \"\"", "pac connection list --environment \"\"", "pac copilot pull --project-dir \"\"", "pac copilot push --project-dir \"\"", "pac copilot publish --bot \"\" --environment \"\"" ], "acceptance_checks": [ "Every exposed operation has one persona-language case and one SKILL.md.", "Exactly two synthetic knowledge files are present.", "Responses state the relevant human approval or review boundary.", "No operation claims an external write, communication, commitment, or protected decision.", "Publishing remains a separate user-approved step." ], "hard_mode": [ "Run pac copilot init and verify settings.mcs.yml before authoring components.", "Author global instructions that preserve the packaged synthetic-only and human-approval boundaries.", "Add the two knowledge files and one skill for every exposed operation.", "Create dedicated connection references only for approved systems and verify least-privilege access.", "Pull before push, inspect every generated file, replay every locked case, and stop before publish." ] }, "curated_copy": true, "source": "field", "origin": "both", "onepager": "69. Campaign Delivery Optimizer Agent one-pager.pptx", "demo_video": "69-Campaign Delivery Optimizer Agent.mp4", "has_onepager": true, "has_demo_video": true, "promise_coverage": 100, "is_primary": true, "aliases": [], "package": { "slug": "campaign-delivery-optimizer", "quest_url": "solutions/campaign-delivery-optimizer/quest.html", "manual_tutorial_url": "solutions/campaign-delivery-optimizer/manual-tutorial.html", "export_manifest_url": "solutions/campaign-delivery-optimizer/export-manifest.json" } }, "_demo": { "slug": "campaign-delivery-optimizer", "case_count": 3, "personas": [ "Ad Operations Manager", "Ad Sales Lead", "Campaign Manager" ], "prompts": [ "Is the Harborlight Streaming campaign falling behind this week, and by how much?", "Where can we find extra reach this week to close the Harborlight gap?", "Draft the make-good for Harborlight so Maya can review it. Don't send anything to the advertiser yet." ] }, "_catalog_kind": "solution", "_readiness": [ "Demo proven", "Conversation tested", "Synthetic data", "Production pattern" ] }, { "schema": "rapp-agent/1.0", "name": "@aibast-agents-library/content-launch-clearance", "version": "1.0.0", "display_name": "Content Launch Clearance Agent", "description": "Help programming, distribution and rights teams confirm where, when and on which platform a title can launch from a synthetic rights snapshot: the launch clearance by territory and platform with the exact contract clause cited, the distribution window and any holdback that blocks an earlier date, and whether the request is ready for rights-team approval. The agent never schedules or publishes a title, changes a contract or submits an approval; the rights team approves every launch through future authenticated tools.", "author": "AIBAST", "tags": [ "media", "streaming", "rights-and-clearances", "content-distribution", "launch-planning" ], "category": "telecom_media_gaming", "quality_tier": "verified", "requires_env": [], "dependencies": [ "@rapp/basic-agent" ], "_file": "agents/@aibast-agents-library/telecom_media_gaming_stacks/content_launch_clearance_stack/content_launch_clearance_agent.py", "_size_kb": 13.6, "_lines": 267, "_sha256": "04793d0bb0e7c6e8f6244bacfd81ae972391e34db6dba5d060ae027da1f36f37", "_install_prefix": "content_launch_clearance__", "_install_filename": "content_launch_clearance__04793d0bb0e7_agent.py", "_stack": "content_launch_clearance", "_stack_vertical": "telecom_media_gaming", "_synthetic_data": true, "_added_at": "2026-10-23T10:00:37-04:00", "_solution": { "advertised_name": "Content Launch Clearance Agent", "sharepoint_list_name": "Content Launch Clearance Agent", "slot": 68, "executive_summary": "Help programming, distribution, and rights teams confirm where, when, and on which platform a title can launch by citing the exact contract clause behind each clearance answer, showing the distribution window and any holdback that blocks an earlier date, and checking what is still missing before rights approval, while every approval stays with the rights team.", "industries": [ "Media & Entertainment", "Streaming", "Broadcasting" ], "personas": [ "Programming Manager", "Distribution Lead", "Rights & Clearances Lead" ], "featured_tools": [ "SharePoint", "Power Automate", "Microsoft Teams" ], "agent_requirements": [ "Microsoft Copilot Studio" ], "capabilities": [ "Confirm where, when, and on which platforms each title can launch", "Cite the exact contract clause behind every clearance answer", "Flag holdbacks that block earlier launch dates", "Show which approvals are still needed before launch" ], "outcomes": [ "Brought titles to audiences faster", "Confirmed launch dates with the contract terms cited", "Caught holdbacks before they became launch problems", "Freed rights teams to focus on deals, not lookups" ], "customer_scenario": [ "Rights questions meant hours of searching through long contracts", "Holdbacks and territory limits were easy to miss under deadline", "Launch plans stalled while teams waited on rights answers", "Approval status lived in email threads, not one shared view" ], "slide_title": "Content Launch Clearance Agent", "slide_summary": "Confirms in seconds where, when, and on which platforms a title can launch, citing the exact contract terms behind each answer.", "scenario_name": "A global media company", "challenge_intro": "A global media company struggled to confirm launch rights quickly:", "actions_intro": "The media company leveraged the Content Launch Clearance Agent to:", "outcomes_intro": "With the agent\u2019s help, the media company:", "opportunity_statements": [ "Assess launch clearance by territory and platform, with the exact contract clause cited for every answer", "Retrieve distribution windows and holdbacks so teams see which launch dates are open and why", "Check approval readiness before launch so no title goes live without its sign-off" ], "module_name": null, "module_description": null, "source_verified": true, "sales_headline": "Confirm where and when every title can launch without handing rights approval to a machine", "card_pitch": "Give programming, distribution, and rights teams one conversational view of each title's launch clearance, contract clause, distribution window, holdbacks, and approval readiness, while every approval stays with the rights team.", "why_try": "The demo lets a programming manager ask a natural question about a title and receive a clearance answer with the contract clause cited, with the rights-approval boundary visible in the response.", "customer_challenge_copy": "Media companies lose launch time searching long contracts for territory and platform rights, and holdbacks surface late while approval status sits in email threads.", "microsoft_ai_story": "Copilot Studio and Microsoft Teams give programming and rights teams a conversational assistant; approved contract files in SharePoint and approval routing in Power Automate can be bound through least-privilege Power Platform connections with audit controls.", "business_value": [ "Shortens the time to confirm a title's launch rights with the clause cited.", "Prevents launches that breach a holdback or territory limit.", "Gives one shared view of what each launch still needs before approval." ], "search_terms": [ "Content Launch Clearance Agent", "Distribution Lead", "Microsoft Copilot Studio", "Microsoft SharePoint", "Microsoft Teams", "Power Automate", "Programming Manager", "Rights & Clearances Lead", "approval_readiness", "content rights", "distribution window", "distribution_window", "holdback", "launch clearance", "launch_clearance", "streaming launch", "territory rights" ], "journey_stage": "Optimize", "blueprint_role": "Separates clearance research and approval-request drafting from protected scheduling, publishing, contract, and rights-approval decisions.", "sample_prompts": [ { "label": "Programming Manager: Launch clearance", "prompt": "Can we put Silver Harbor on our streaming service in the UK, and which contract clause says so?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=content-launch-clearance-launch-clearance" }, { "label": "Distribution Lead: Distribution window and holdback", "prompt": "Programming wants Paper Lantern live in Canada in about a month. Is there a holdback that blocks that date?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=content-launch-clearance-distribution-window" }, { "label": "Rights & Clearances Lead: Approval readiness draft", "prompt": "Is Northern Signal ready to come to the rights team for sign-off? What is still missing?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=content-launch-clearance-approval-readiness" } ], "architecture": { "business_flow": [ "Programming managers, distribution leads, and rights teams", "Microsoft Teams or Microsoft 365 Copilot", "Content Launch Clearance Agent", "Approved contract files in Microsoft SharePoint", "Distribution windows and rights approval status", "Authorized rights-team approval workflow" ], "capabilities": [ { "operation": "launch_clearance", "name": "Launch clearance", "purpose": "Use when someone asks whether a title can launch in a territory on the owned streaming service, or which contract clause supports the answer." }, { "operation": "distribution_window", "name": "Distribution window and holdback", "purpose": "Use when someone asks when a title's rights window opens, which holdback applies, or whether a requested launch date is open." }, { "operation": "approval_readiness", "name": "Approval readiness draft", "purpose": "Use when someone asks whether a launch request is ready for rights-team approval or what is still missing; returns an approval request draft that is never submitted." } ], "easy_mode": [ "Verify Power Platform CLI 2.9.3 or newer and authenticate to the approved target environment.", "Initialize a modern cli-copilot project with an approved publisher prefix and preserve the generated identity settings.", "Upload the two packaged synthetic knowledge files and every operation-specific SKILL.md.", "Bind only approved least-privilege production connections after security, privacy, and business-owner review.", "Push the Draft agent, replay every locked persona case, and stop before publish for explicit approval." ], "local_install_prompt": "Install Content Launch Clearance Agent from https://raw.githubusercontent.com/microsoft/aibast-agents-library/main/solutions/content-launch-clearance/deployment.json, verify the expected tool and smoke case, and preserve the synthetic read-only boundary. Do not ask me to open a terminal. Do not connect customer systems or execute external actions.", "copilot_studio_prompt": "Using the Microsoft Copilot Studio plugin, initialize a Draft agent for Content Launch Clearance Agent from https://raw.githubusercontent.com/microsoft/aibast-agents-library/main/solutions/content-launch-clearance/deployment.json. Upload the two synthetic knowledge files and 3 skills, replay every locked case, and Stop before publish. Do not bind production connections without explicit customer approval.", "required_connections": [ "Use only approved, least-privilege production connections; the packaged demonstration uses synthetic knowledge files.", "Approved rights contracts in Microsoft SharePoint (read-only)", "Distribution windows and holdback records (read-only)", "Rights approval status in Power Automate approvals (read-only)" ], "manual_commands": [ "pac auth create", "pac copilot init --name \"Content Launch Clearance Agent\" --publisher-prefix --authoring-mode cli-copilot --project-dir \"\" --environment \"\"", "pac connection list --environment \"\"", "pac copilot pull --project-dir \"\"", "pac copilot push --project-dir \"\"", "pac copilot publish --bot \"\" --environment \"\"" ], "acceptance_checks": [ "Every exposed operation has one persona-language case and one SKILL.md.", "Exactly two synthetic knowledge files are present.", "Responses state the relevant human approval or review boundary.", "No operation claims an external write, communication, commitment, or protected decision.", "Publishing remains a separate user-approved step." ], "hard_mode": [ "Run pac copilot init and verify settings.mcs.yml before authoring components.", "Author global instructions that preserve the packaged synthetic-only and human-approval boundaries.", "Add the two knowledge files and one skill for every exposed operation.", "Create dedicated connection references only for approved systems and verify least-privilege access.", "Pull before push, inspect every generated file, replay every locked case, and stop before publish." ] }, "curated_copy": true, "source": "field", "origin": "both", "onepager": "68. Content Launch Clearance Agent one-pager.pptx", "demo_video": "68-Content Launch Clearance Agent.mp4", "has_onepager": true, "has_demo_video": true, "promise_coverage": 100, "is_primary": true, "aliases": [], "package": { "slug": "content-launch-clearance", "quest_url": "solutions/content-launch-clearance/quest.html", "manual_tutorial_url": "solutions/content-launch-clearance/manual-tutorial.html", "export_manifest_url": "solutions/content-launch-clearance/export-manifest.json" } }, "_demo": { "slug": "content-launch-clearance", "case_count": 3, "personas": [ "Distribution Lead", "Programming Manager", "Rights & Clearances Lead" ], "prompts": [ "Can we put Silver Harbor on our streaming service in the UK, and which contract clause says so?", "Programming wants Paper Lantern live in Canada in about a month. Is there a holdback that blocks that date?", "Is Northern Signal ready to come to the rights team for sign-off? What is still missing?" ] }, "_catalog_kind": "solution", "_readiness": [ "Demo proven", "Conversation tested", "Synthetic data", "Production pattern" ] }, { "schema": "rapp-agent/1.0", "name": "@aibast-agents-library/first-time-fix", "version": "1.0.0", "display_name": "First-Time Fix Agent", "description": "Help contact center and field service agents resolve a customer's connection issue on the first call from a synthetic broadband snapshot: the service status with the guided connection fix, local network work that could be affecting the service, previous visit outcomes with technician notes, and the recommended technician with the job preparation when a visit is needed. The agent never books, changes or cancels a visit, changes a service, contacts a customer or dispatches a technician; a human agent confirms every booking through future authenticated tools.", "author": "AIBAST", "tags": [ "telecommunications", "broadband", "contact-center", "field-service", "first-call-resolution" ], "category": "telecom_media_gaming", "quality_tier": "verified", "requires_env": [], "dependencies": [ "@rapp/basic-agent" ], "_file": "agents/@aibast-agents-library/telecom_media_gaming_stacks/first_time_fix_stack/first_time_fix_agent.py", "_size_kb": 14.9, "_lines": 303, "_sha256": "35c333da8a89eb74aaaebedee0d92840329b02cb0b7e6d31cd29832f26ee86c4", "_install_prefix": "first_time_fix__", "_install_filename": "first_time_fix__35c333da8a89_agent.py", "_stack": "first_time_fix", "_stack_vertical": "telecom_media_gaming", "_synthetic_data": true, "_added_at": "2026-10-23T10:00:37-04:00", "_solution": { "advertised_name": "First-Time Fix Agent", "sharepoint_list_name": "First-Time Fix Agent", "slot": 67, "executive_summary": "Help contact center and field service teams resolve connection issues on the first call by guiding the right fix for each customer's service, checking local network work and past visits before a technician is sent, and preparing the technician's job when a visit is needed, while every booking stays with a person to confirm.", "industries": [ "Telecommunications", "Broadband & Cable" ], "personas": [ "Contact Center Agent", "Field Service Dispatcher" ], "featured_tools": [ "Dynamics 365 Customer Service", "Dynamics 365 Field Service", "SharePoint" ], "agent_requirements": [ "Microsoft Copilot Studio" ], "capabilities": [ "Guide agents through the right fix for each customer's service", "Check local work and past visits before sending a technician", "Prepare each job with the fault history and the parts to bring", "Keep a person in control of every technician booking" ], "outcomes": [ "Solved more connection issues on the first call", "Avoided unneeded truck rolls during area outages", "Sent technicians who arrived prepared for the fault", "Put technicians where they add the most value" ], "customer_scenario": [ "Agents lacked one view of service, local work, and visit history", "Technicians were sent for issues a guided fix or an outage explained", "Technicians arrived without knowing what had already been tried", "Repeat visits raised costs and frustrated customers" ], "slide_title": "First-Time Fix Agent", "slide_summary": "Guides connection fixes on the first call and prepares the right technician's job, for a person to confirm, when a visit is needed.", "scenario_name": "A regional broadband provider", "challenge_intro": "A regional broadband provider struggled to resolve connection issues:", "actions_intro": "The broadband provider leveraged the First-Time Fix Agent to:", "outcomes_intro": "With the agent\u2019s help, the broadband provider:", "opportunity_statements": [ "Guide the connection fix from the customer's service, local work, and past visits to solve more on the first call", "Check local work and past visits before any visit to avoid truck rolls for area outages and repeat faults", "Prepare the right technician's job with the fault history and parts to bring, ready for a person to confirm" ], "module_name": null, "module_description": null, "source_verified": true, "sales_headline": "Solve more connection issues on the first call without handing bookings to a machine", "card_pitch": "Give contact center and field service teams one conversational view of the customer's service, the guided fix, local network work, past visits, and the right technician's job, while every booking stays with a person to confirm.", "why_try": "The demo lets a contact center agent ask a natural question mid-call and receive a source-backed guided fix, with the booking boundary visible in the response.", "customer_challenge_copy": "Contact centers send technicians for issues a guided fix or a local outage already explains, and technicians arrive without the history of what was tried.", "microsoft_ai_story": "Copilot Studio and Microsoft Teams give agents a conversational assistant during the call; Dynamics 365 Customer Service, Dynamics 365 Field Service, and SharePoint can be bound through least-privilege Power Platform connections with audit controls.", "business_value": [ "Raises first-call resolution with a guided fix for each service.", "Avoids truck rolls that local network work or a known fault explains.", "Sends technicians prepared with the fault history and parts." ], "search_terms": [ "Contact Center Agent", "Dynamics 365 Customer Service", "Dynamics 365 Field Service", "Field Service Dispatcher", "First-Time Fix Agent", "Microsoft Copilot Studio", "Microsoft SharePoint", "broadband", "connection issue", "first-call resolution", "guided_fix", "local_work_check", "network outage", "technician booking", "technician_booking", "truck roll", "visit_history" ], "journey_stage": "Optimize", "blueprint_role": "Separates evidence gathering and job preparation from protected booking, dispatch, and customer-contact decisions.", "sample_prompts": [ { "label": "Contact Center Agent: Guided connection fix", "prompt": "The customer on Alder Grove Service keeps dropping. What guided fix should I walk them through right now?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=first-time-fix-guided-fix" }, { "label": "Contact Center Agent: Local work check", "prompt": "Is there any network work near Meadow Lane that could explain the outage?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=first-time-fix-local-work-check" }, { "label": "Field Service Dispatcher: Previous visit review", "prompt": "What did the last technician find at Alder Grove Service?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=first-time-fix-visit-history" }, { "label": "Field Service Dispatcher: Technician booking draft", "prompt": "If the light stays red, which technician should go to Alder Grove and what should they prepare? Don't book it yet.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=first-time-fix-technician-booking" } ], "architecture": { "business_flow": [ "Contact center agents and field service dispatchers", "Microsoft Teams or Microsoft 365 Copilot", "First-Time Fix Agent", "Dynamics 365 Customer Service and service status", "Network work and Dynamics 365 Field Service visit history", "Authorized agent booking workflow" ], "capabilities": [ { "operation": "guided_fix", "name": "Guided connection fix", "purpose": "Use when a contact center agent asks which fix to walk the customer through during the call, or what the service status is." }, { "operation": "local_work_check", "name": "Local work check", "purpose": "Use when the agent asks whether planned or emergency network work near the service could explain the issue." }, { "operation": "visit_history", "name": "Previous visit review", "purpose": "Use when someone asks what earlier technician visits found or what the technician notes say." }, { "operation": "technician_booking", "name": "Technician booking draft", "purpose": "Use when someone asks which technician to send and what job preparation is needed; returns a booking draft that is never booked." } ], "easy_mode": [ "Verify Power Platform CLI 2.9.3 or newer and authenticate to the approved target environment.", "Initialize a modern cli-copilot project with an approved publisher prefix and preserve the generated identity settings.", "Upload the two packaged synthetic knowledge files and every operation-specific SKILL.md.", "Bind only approved least-privilege production connections after security, privacy, and business-owner review.", "Push the Draft agent, replay every locked persona case, and stop before publish for explicit approval." ], "local_install_prompt": "Install First-Time Fix Agent from https://raw.githubusercontent.com/microsoft/aibast-agents-library/main/solutions/first-time-fix/deployment.json, verify the expected tool and smoke case, and preserve the synthetic read-only boundary. Do not ask me to open a terminal. Do not connect customer systems or execute external actions.", "copilot_studio_prompt": "Using the Microsoft Copilot Studio plugin, initialize a Draft agent for First-Time Fix Agent from https://raw.githubusercontent.com/microsoft/aibast-agents-library/main/solutions/first-time-fix/deployment.json. Upload the two synthetic knowledge files and 4 skills, replay every locked case, and Stop before publish. Do not bind production connections without explicit customer approval.", "required_connections": [ "Use only approved, least-privilege production connections; the packaged demonstration uses synthetic knowledge files.", "Dynamics 365 Customer Service cases and service status (read-only)", "Network maintenance and outage records (read-only)", "Dynamics 365 Field Service visits and technician schedules (read-only)" ], "manual_commands": [ "pac auth create", "pac copilot init --name \"First-Time Fix Agent\" --publisher-prefix --authoring-mode cli-copilot --project-dir \"\" --environment \"\"", "pac connection list --environment \"\"", "pac copilot pull --project-dir \"\"", "pac copilot push --project-dir \"\"", "pac copilot publish --bot \"\" --environment \"\"" ], "acceptance_checks": [ "Every exposed operation has one persona-language case and one SKILL.md.", "Exactly two synthetic knowledge files are present.", "Responses state the relevant human approval or review boundary.", "No operation claims an external write, communication, commitment, or protected decision.", "Publishing remains a separate user-approved step." ], "hard_mode": [ "Run pac copilot init and verify settings.mcs.yml before authoring components.", "Author global instructions that preserve the packaged synthetic-only and human-approval boundaries.", "Add the two knowledge files and one skill for every exposed operation.", "Create dedicated connection references only for approved systems and verify least-privilege access.", "Pull before push, inspect every generated file, replay every locked case, and stop before publish." ] }, "curated_copy": true, "source": "field", "origin": "both", "onepager": "67. First-Time Fix Agent one-pager.pptx", "demo_video": "67-First-Time Fix Agent.mp4", "has_onepager": true, "has_demo_video": true, "promise_coverage": 100, "is_primary": true, "aliases": [], "package": { "slug": "first-time-fix", "quest_url": "solutions/first-time-fix/quest.html", "manual_tutorial_url": "solutions/first-time-fix/manual-tutorial.html", "export_manifest_url": "solutions/first-time-fix/export-manifest.json" } }, "_demo": { "slug": "first-time-fix", "case_count": 4, "personas": [ "Contact Center Agent", "Field Service Dispatcher" ], "prompts": [ "The customer on Alder Grove Service keeps dropping. What guided fix should I walk them through right now?", "Is there any network work near Meadow Lane that could explain the outage?", "What did the last technician find at Alder Grove Service?" ] }, "_catalog_kind": "solution", "_readiness": [ "Demo proven", "Conversation tested", "Synthetic data", "Production pattern" ] }, { "schema": "rapp-agent/1.0", "name": "@aibast-agents-library/live-event-launch-planner", "version": "1.0.0", "display_name": "Live Event Launch Planner Agent", "description": "Help live operations, community and marketing teams launch a seasonal in-game event from one plan built on a synthetic snapshot: the launch timeline with every milestone to launch day, the readiness checklist across teams with owners and status, the player announcement with its approval and publishing permission, and a summary of late, missing and blocked work. The agent never publishes an announcement, changes a task, sends a reminder or changes the event; the event owners approve every step through future authenticated tools.", "author": "AIBAST", "tags": [ "gaming", "live-operations", "community", "event-launch", "readiness" ], "category": "telecom_media_gaming", "quality_tier": "verified", "requires_env": [], "dependencies": [ "@rapp/basic-agent" ], "_file": "agents/@aibast-agents-library/telecom_media_gaming_stacks/live_event_launch_planner_stack/live_event_launch_planner_agent.py", "_size_kb": 14.4, "_lines": 277, "_sha256": "9503f99b9605e77a3b07af943712d434c67b085b0e48d858c2ff7b5424c2a50b", "_install_prefix": "live_event_launch_planner__", "_install_filename": "live_event_launch_planner__9503f99b9605_agent.py", "_stack": "live_event_launch_planner", "_stack_vertical": "telecom_media_gaming", "_synthetic_data": true, "_added_at": "2026-10-23T10:00:37-04:00", "_solution": { "advertised_name": "Live Event Launch Planner Agent", "sharepoint_list_name": "Live Event Launch Planner Agent", "slot": 70, "executive_summary": "Help live operations, community, and marketing teams launch seasonal in-game events from one plan by building the timeline from communications approval to launch day, compiling readiness across teams with owners and status, and checking player announcements before publishing, while nothing is published or sent without a person's approval.", "industries": [ "Gaming", "Interactive Media" ], "personas": [ "Live Operations Manager", "Community Manager", "Marketing Lead" ], "featured_tools": [ "Microsoft Planner", "Microsoft Teams", "SharePoint, Outlook" ], "agent_requirements": [ "Microsoft Copilot Studio" ], "capabilities": [ "Build each event's timeline, from approvals to launch day", "Compile one readiness checklist across live ops and marketing", "Draft player announcements and check publishing permission", "Summarize launch status for every team in one place" ], "outcomes": [ "Delivered smoother seasonal event launches", "Drew bigger player turnout on launch day", "Kept teams aligned without extra status meetings", "Published announcements only once approved" ], "customer_scenario": [ "Launch plans lived in scattered documents and chat threads", "Teams relied on status meetings to see what was ready", "Late approvals put announcement dates at risk", "Players heard about events late or in mixed messages" ], "slide_title": "Live Event Launch Planner Agent", "slide_summary": "Builds the launch plan for a seasonal in-game event: a timeline, readiness checks across teams, and draft player announcements.", "scenario_name": "A global game studio", "challenge_intro": "A global game studio struggled to coordinate seasonal event launches:", "actions_intro": "The game studio leveraged the Live Event Launch Planner Agent to:", "outcomes_intro": "With the agent\u2019s help, the game studio:", "opportunity_statements": [ "Build the launch timeline with every milestone, from communications approval to launch day", "Compile readiness across teams so live ops, community, and marketing see what is still open", "Check and draft player announcements and hold them until publishing is approved" ], "module_name": null, "module_description": null, "source_verified": true, "sales_headline": "Launch every seasonal event from one plan without handing publishing to a machine", "card_pitch": "Give live operations, community, and marketing teams one conversational view of each seasonal event's timeline, cross-team readiness, player announcement, and open work, while every announcement and reminder stays with a person to approve.", "why_try": "The demo lets a live operations manager ask a natural question about an upcoming event and receive its timeline, open tasks with owners, and the held player announcement, with the publishing boundary visible in the response.", "customer_challenge_copy": "Game studios coordinate seasonal event launches across scattered documents and status meetings, and late approvals put player announcements at risk.", "microsoft_ai_story": "Copilot Studio and Microsoft Teams give live operations, community, and marketing teams a conversational assistant; launch tasks, Teams updates, approved SharePoint launch materials, and Outlook drafts can be bound through least-privilege Power Platform connections with audit controls.", "business_value": [ "Builds each event's timeline from approvals to launch day in one place.", "Surfaces late, missing, and blocked work with its owner before launch day.", "Holds player announcements until publishing is approved." ], "search_terms": [ "Community Manager", "Live Event Launch Planner Agent", "Live Operations Manager", "Marketing Lead", "Microsoft Copilot Studio", "Microsoft Outlook", "Microsoft Planner", "Microsoft SharePoint", "Microsoft Teams", "announcement_check", "game launch", "launch readiness", "launch_summary", "launch_timeline", "live operations", "live service", "player announcement", "readiness_checklist", "seasonal event" ], "journey_stage": "Optimize", "blueprint_role": "Separates launch planning, readiness tracking, and announcement drafting from protected publishing, reminder, and event-configuration decisions.", "sample_prompts": [ { "label": "Live Operations Manager: Launch timeline", "prompt": "Lay out the Frostveil Festival plan from comms approval to launch day. What comes first?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=live-event-launch-planner-launch-timeline" }, { "label": "Live Operations Manager: Cross-team readiness checklist", "prompt": "Who owns each task across live ops, community and marketing for Frostveil, and where does each one stand?", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=live-event-launch-planner-readiness-checklist" }, { "label": "Community Manager: Player announcement check", "prompt": "Can we post the Frostveil player announcement yet? Show me the copy but hold it.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=live-event-launch-planner-announcement-check" }, { "label": "Marketing Lead: Launch status summary", "prompt": "What's late or blocked for Frostveil, and who needs a nudge? Don't message anyone yet.", "demo_url": "solutions/_shared/m365-copilot-demo.html?scenario=live-event-launch-planner-launch-summary" } ], "architecture": { "business_flow": [ "Live operations managers, community managers, and marketing leads", "Microsoft Teams or Microsoft 365 Copilot", "Live Event Launch Planner Agent", "Launch tasks and Microsoft Teams updates", "Approved launch materials in Microsoft SharePoint", "Authorized publishing approval workflow" ], "capabilities": [ { "operation": "launch_timeline", "name": "Launch timeline", "purpose": "Use when someone asks for an event's milestones from communications approval to launch day, or what comes first." }, { "operation": "readiness_checklist", "name": "Cross-team readiness checklist", "purpose": "Use when someone asks which launch tasks each team owns, who owns them, and where each one stands." }, { "operation": "announcement_check", "name": "Player announcement check", "purpose": "Use when someone asks whether the player announcement is approved and may be published, or wants to see the copy; returns a draft that is never published." }, { "operation": "launch_summary", "name": "Launch status summary", "purpose": "Use when someone asks what is late, missing, or blocked and who needs a reminder; returns reminder drafts that are never sent." } ], "easy_mode": [ "Verify Power Platform CLI 2.9.3 or newer and authenticate to the approved target environment.", "Initialize a modern cli-copilot project with an approved publisher prefix and preserve the generated identity settings.", "Upload the two packaged synthetic knowledge files and every operation-specific SKILL.md.", "Bind only approved least-privilege production connections after security, privacy, and business-owner review.", "Push the Draft agent, replay every locked persona case, and stop before publish for explicit approval." ], "local_install_prompt": "Install Live Event Launch Planner Agent from https://raw.githubusercontent.com/microsoft/aibast-agents-library/main/solutions/live-event-launch-planner/deployment.json, verify the expected tool and smoke case, and preserve the synthetic read-only boundary. Do not ask me to open a terminal. Do not connect customer systems or execute external actions.", "copilot_studio_prompt": "Using the Microsoft Copilot Studio plugin, initialize a Draft agent for Live Event Launch Planner Agent from https://raw.githubusercontent.com/microsoft/aibast-agents-library/main/solutions/live-event-launch-planner/deployment.json. Upload the two synthetic knowledge files and 4 skills, replay every locked case, and Stop before publish. Do not bind production connections without explicit customer approval.", "required_connections": [ "Use only approved, least-privilege production connections; the packaged demonstration uses synthetic knowledge files.", "Launch plan tasks and owners (read-only)", "Microsoft Teams launch channel updates (read-only)", "Approved launch materials and announcement briefs in Microsoft SharePoint (read-only)" ], "manual_commands": [ "pac auth create", "pac copilot init --name \"Live Event Launch Planner Agent\" --publisher-prefix --authoring-mode cli-copilot --project-dir \"\" --environment \"\"", "pac connection list --environment \"\"", "pac copilot pull --project-dir \"\"", "pac copilot push --project-dir \"\"", "pac copilot publish --bot \"\" --environment \"\"" ], "acceptance_checks": [ "Every exposed operation has one persona-language case and one SKILL.md.", "Exactly two synthetic knowledge files are present.", "Responses state the relevant human approval or review boundary.", "No operation claims an external write, communication, commitment, or protected decision.", "Publishing remains a separate user-approved step." ], "hard_mode": [ "Run pac copilot init and verify settings.mcs.yml before authoring components.", "Author global instructions that preserve the packaged synthetic-only and human-approval boundaries.", "Add the two knowledge files and one skill for every exposed operation.", "Create dedicated connection references only for approved systems and verify least-privilege access.", "Pull before push, inspect every generated file, replay every locked case, and stop before publish." ] }, "curated_copy": true, "source": "field", "origin": "both", "onepager": "70. Live Event Launch Planner Agent one-pager.pptx", "demo_video": "70-Live Event Launch Planner Agent.mp4", "has_onepager": true, "has_demo_video": true, "promise_coverage": 100, "is_primary": true, "aliases": [], "package": { "slug": "live-event-launch-planner", "quest_url": "solutions/live-event-launch-planner/quest.html", "manual_tutorial_url": "solutions/live-event-launch-planner/manual-tutorial.html", "export_manifest_url": "solutions/live-event-launch-planner/export-manifest.json" } }, "_demo": { "slug": "live-event-launch-planner", "case_count": 4, "personas": [ "Community Manager", "Live Operations Manager", "Marketing Lead" ], "prompts": [ "Lay out the Frostveil Festival plan from comms approval to launch day. What comes first?", "Who owns each task across live ops, community and marketing for Frostveil, and where does each one stand?", "Can we post the Frostveil player announcement yet? Show me the copy but hold it." ] }, "_catalog_kind": "solution", "_readiness": [ "Demo proven", "Conversation tested", "Synthetic data", "Production pattern" ] }, { "schema": "rapp-agent/1.0", "name": "@aibast-agents-library/copilot-studio-deploy", "version": "0.1.0", "display_name": "Deploy to Copilot Studio", "description": "Deploys an AIBAST library solution into the user's own Copilot Studio environment with their own sign-in: checks first, imports, publishes, and can undo.", "author": "AIBAST", "tags": [ "copilot-studio", "deploy", "power-platform" ], "category": "general", "quality_tier": "pilot", "requires_env": [], "dependencies": [ "@rapp/basic-agent" ], "_file": "agents/@aibast-agents-library/templates/copilot_studio_deploy_agent.py", "_size_kb": 13.1, "_lines": 294, "_sha256": "fe3046c1372ee78812bddd29fba26d159c798916960ca9ab68d855e598eb4095", "_install_prefix": "copilot_studio_deploy__", "_install_filename": "copilot_studio_deploy__fe3046c1372e_agent.py", "_stack": null, "_stack_vertical": null, "_synthetic_data": false, "_added_at": "2026-11-05T20:43:45-05:00", "_catalog_kind": "template", "_readiness": [] }, { "schema": "rapp-agent/1.0", "name": "@aibast-agents-library/industry-agent-generator", "version": "1.0.0", "display_name": "Industry Agent Generator", "description": "Generates a complete, constitution-compliant AIBAST industry solution agent from a one-pager brief, with its locked demo cases.", "author": "AIBAST", "tags": [ "meta", "generator", "scaffolding", "constitution", "agent-authoring" ], "category": "general", "quality_tier": "verified", "requires_env": [], "dependencies": [ "@rapp/basic-agent" ], "_file": "agents/@aibast-agents-library/templates/industry_agent_generator_agent.py", "_size_kb": 31.0, "_lines": 695, "_sha256": "df4e721bccc8fefc6333540868248afb97a390ee65e76da81407bad2522c0d97", "_install_prefix": "industry_agent_generator__", "_install_filename": "industry_agent_generator__df4e721bccc8_agent.py", "_stack": null, "_stack_vertical": null, "_synthetic_data": true, "_added_at": "2026-08-07T18:32:19-04:00", "_catalog_kind": "template", "_readiness": [ "Synthetic data" ] }, { "schema": "rapp-agent/1.0", "name": "@aibast-agents-library/workshop", "version": "1.0.0", "display_name": "AIBAST Workshop Engine", "description": "Discovers any packaged AIBAST solution, hot-loads and tests its business agent, prepares its Copilot Studio Draft, and validates the real Preview evidence.", "author": "AIBAST", "tags": [ "workshop", "personless-harness", "brainstem", "copilot-studio", "generic" ], "category": "general", "quality_tier": "pilot", "requires_env": [], "dependencies": [ "@rapp/basic-agent" ], "_file": "agents/@aibast-agents-library/templates/workshop_agent.py", "_size_kb": 56.5, "_lines": 1493, "_sha256": "b852fc7239c8255775afd3ed7986e7d87a01ad11dfc70edcdcec833f22adc134", "_install_prefix": "workshop__", "_install_filename": "workshop__b852fc7239c8_agent.py", "_stack": null, "_stack_vertical": null, "_synthetic_data": false, "_added_at": "2026-08-08T20:37:40-04:00", "_catalog_kind": "template", "_readiness": [] } ], "first_party": [ { "id": "ai-powered-data-enrichment", "name": "AI-powered Data Enrichment for opportunities", "deck_label": "AI-powered Data Enrichment", "product": "Dynamics 365 Sales", "vertical": "b2b_sales", "deck_status": "GA", "deck_description": "Enriches opportunity data from email interactions", "description": "Analyzes recent email and Teams meeting context to suggest or make configured updates to opportunity records.", "use_cases": [ "Analyze recent email exchanges and Teams meeting context for selected opportunities", "Identify missing or outdated opportunity information", "Suggest or automatically make configured opportunity-field updates", "Associate relevant email conversations with the opportunity record" ], "personas": [ "Seller", "Sales Manager", "Sales Operations" ], "overview_url": "https://learn.microsoft.com/en-us/dynamics365/sales/data-enrichment-agent-overview", "configure_url": "https://learn.microsoft.com/en-us/dynamics365/sales/data-enrichment-agent-configure", "doc_status": "dedicated", "source_slide": 1, "_catalog_kind": "first_party", "_documented": true, "_availability_label": "Available" }, { "id": "recommended-actions-agent", "name": "Recommended Actions agent", "product": "Dynamics 365 Sales", "vertical": "b2b_sales", "deck_status": "Preview", "lifecycle": "Preview", "lifecycle_source_url": "https://learn.microsoft.com/en-us/dynamics365/sales/recommended-actions-overview", "deck_description": "Prioritized seller actions based on opportunity context", "description": "Presents prioritized recommended actions so sellers can focus on the most relevant tasks for their opportunities.", "use_cases": [ "Review a prioritized list of recommended seller actions", "Use opportunity context to focus attention on the most relevant next tasks", "Combine recommendations from Sales Opportunity Agent and configured custom agents", "Review Data Enrichment actions alongside the prioritized recommendations" ], "personas": [ "Seller", "Sales Manager" ], "overview_url": "https://learn.microsoft.com/en-us/dynamics365/sales/recommended-actions-overview", "configure_url": "https://learn.microsoft.com/en-us/dynamics365/sales/configure-recommended-actions", "doc_status": "dedicated", "source_slide": 1, "_catalog_kind": "first_party", "_documented": true, "_availability_label": "Preview" }, { "id": "sales-development-agent", "name": "Sales Development agent", "deck_label": "Sales Deployment Agent", "product": "Microsoft Agent 365", "vertical": "b2b_sales", "deck_status": "Preview", "lifecycle": "Preview", "lifecycle_source_url": "https://learn.microsoft.com/en-us/microsoft-agent-365/use", "deck_description": "Automats sales outreach and prospect management within Microsoft Teams", "description": "Automates sales outreach and prospect management in Microsoft Teams through an Agent 365 agent with its own identity.", "use_cases": [ "Automate personalized prospect outreach and follow-up in Microsoft Teams", "Configure the agent playbook, guidelines, product knowledge, and settings through chat", "Integrate Dynamics 365 for lead management and activity tracking", "Review prospect status and scale a reviewed configuration across teams" ], "personas": [ "Sales Development Representative", "Sales Manager", "Agent Administrator" ], "overview_url": "https://learn.microsoft.com/en-us/microsoft-agent-365/use", "configure_url": "https://learn.microsoft.com/en-us/microsoft-agent-365/use#deploy-and-create-the-sales-development-agent", "doc_status": "dedicated", "source_slide": 1, "_catalog_kind": "first_party", "_documented": true, "_availability_label": "Preview" }, { "id": "sales-opportunity-agent", "name": "Sales Opportunity Agent", "product": "Dynamics 365 Sales", "vertical": "b2b_sales", "deck_status": "GA", "deck_description": "Continuous opportunity research, risk & stakeholder intelligence", "description": "Researches open opportunities and provides deal prioritization, risk, stakeholder, competitor, and deal-health insights.", "use_cases": [ "Continuously research open opportunities using available sales context", "Surface deal risks and material changes that need seller attention", "Prioritize opportunities so sellers can focus on the deals that need action", "Review stakeholder, competitor, and deal-health context for an opportunity" ], "personas": [ "Seller", "Sales Manager", "Sales Director" ], "overview_url": "https://learn.microsoft.com/en-us/dynamics365/sales/sales-opportunity-agent", "configure_url": "https://learn.microsoft.com/en-us/dynamics365/sales/configure-opportunity-research-agent", "doc_status": "dedicated", "source_slide": 1, "_catalog_kind": "first_party", "_documented": true, "_availability_label": "Available" }, { "id": "sales-qualification-agent", "name": "Sales Qualification Agent", "product": "Dynamics 365 Sales", "vertical": "b2b_sales", "deck_status": "GA", "deck_description": "Researches leads, evaluates fit, autonomous outreach", "description": "Automates lead research and, when configured for research and engage mode, lead outreach and follow-up.", "use_cases": [ "Research selected leads and assess their fit against a target customer profile", "Generate an initial outreach email for matching leads in research-only mode", "Send outreach and follow-up messages in research and engage mode", "Evaluate purchase intent and BANT criteria before handing promising leads to sellers" ], "personas": [ "Sales Development Representative", "Seller", "Sales Supervisor" ], "overview_url": "https://learn.microsoft.com/en-us/dynamics365/sales/sales-qualification-agent", "configure_url": "https://learn.microsoft.com/en-us/dynamics365/sales/configure-sales-qualification-agent", "doc_status": "dedicated", "source_slide": 1, "_catalog_kind": "first_party", "_documented": true, "_availability_label": "Available" }, { "id": "sales-research-agent", "name": "Sales Research Agent", "product": "Dynamics 365 Sales", "vertical": "b2b_sales", "deck_status": "GA", "deck_description": "Natural language dialog with sales data", "description": "Answers strategic sales questions through natural-language research over Dynamics 365 Sales data and other configured data sources.", "use_cases": [ "Explore pipeline and sales operations questions using natural language", "Research Dynamics 365 Sales data together with uploaded CSV, Excel, or PDF files", "Generate a research blueprint with findings, suggested next steps, and visualizations", "Ask follow-up questions, change visualizations, and add data to an active analysis" ], "personas": [ "Seller", "Sales Manager", "Sales Operations" ], "overview_url": "https://learn.microsoft.com/en-us/dynamics365/sales/sales-research-agent", "configure_url": "https://learn.microsoft.com/en-us/dynamics365/sales/configure-sales-research-agent", "doc_status": "dedicated", "source_slide": 1, "_catalog_kind": "first_party", "_documented": true, "_availability_label": "Available" }, { "id": "sales-agent-m365-copilot", "name": "Sales agent", "deck_label": "Sales Agent (M365 Copilot)", "product": "Microsoft 365 Copilot", "vertical": "b2b_sales", "deck_status": "GA", "deck_description": "Search, synthesize & act on sales data in M365 apps", "description": "Provides a natural-language interface in Microsoft 365 Copilot for searching, synthesizing, and acting on connected sales data.", "use_cases": [ "Query connected Dynamics 365 Sales or Salesforce CRM data using natural language", "Synthesize CRM records with eligible email, Teams meeting, and Teams message context", "Prepare account and opportunity summaries before customer engagement", "Draft follow-up communications and actions from the available sales context" ], "personas": [ "Seller", "Sales Manager" ], "overview_url": "https://learn.microsoft.com/en-us/microsoft-sales-copilot/sales-chat-overview", "configure_url": "https://learn.microsoft.com/en-us/microsoft-sales-copilot/set-up-sales-chat", "doc_status": "dedicated", "source_slide": 1, "_catalog_kind": "first_party", "_documented": true, "_availability_label": "Available" }, { "id": "case-management-agent", "name": "Case Management Agent", "product": "Dynamics 365 Customer Service", "vertical": "customer_service", "deck_status": "GA", "deck_description": "Automates case lifecycle - create, update, resolve, close", "description": "Creates cases from supported conversations and updates configured case fields from conversation or email context.", "use_cases": [ "Create cases from chat, digital messaging, and incoming voice conversations", "Update configured case fields when a conversation reaches wrap-up", "Update configured fields from incoming emails associated with a case", "Populate suggested field values when a representative creates a case manually" ], "personas": [ "Customer Service Representative", "Service Supervisor", "Service Administrator" ], "overview_url": "https://learn.microsoft.com/en-us/dynamics365/customer-service/use/use-case-creation-agent", "configure_url": "https://learn.microsoft.com/en-us/dynamics365/customer-service/administer/set-up-autonomous-case-agents", "doc_status": "dedicated", "source_slide": 2, "_catalog_kind": "first_party", "_documented": true, "_availability_label": "Available" }, { "id": "customer-intent-agent", "name": "Customer Intent Agent", "product": "Dynamics 365 Contact Center / Customer Service", "vertical": "customer_service", "deck_status": "GA", "deck_description": "Discovers intents from cases/conversations, builds intent library", "description": "Discovers intents from historical customer interactions and manages them as an intent library for assisted and self-service experiences.", "use_cases": [ "Discover new customer intents from historical cases and conversations", "Manage intents, intent groups, and optional lines of business", "Provide representatives with intent-based questions and suggested solutions", "Use intent context to support self-service follow-up questions and knowledge retrieval" ], "personas": [ "Service Administrator", "Contact Center Manager", "Customer Service Representative" ], "overview_url": "https://learn.microsoft.com/en-us/dynamics365/contact-center/administer/overview-customer-intent-agent", "configure_url": "https://learn.microsoft.com/en-us/dynamics365/contact-center/administer/manage-customer-intent-agent", "doc_status": "dedicated", "source_slide": 2, "_catalog_kind": "first_party", "_documented": true, "_availability_label": "Available" }, { "id": "customer-intent-agent-voice", "name": "Customer Intent Agent for voice", "deck_label": "Customer Intent Agent (Voice)", "product": "Dynamics 365 Contact Center", "vertical": "customer_service", "deck_status": "GA", "deck_description": "Autonomous voice resolution using intent library + playbooks", "description": "Uses the Customer Intent Agent intent library to support intent-driven voice self-service and escalation.", "use_cases": [ "Detect customer intent during an inbound voice interaction", "Ask intent-specific follow-up questions and provide a solution from configured resources", "Escalate the conversation to a representative when self-service cannot resolve it", "Connect an intent-enabled voice agent to a queue or workstream" ], "personas": [ "Contact Center Administrator", "Contact Center Manager", "Customer Service Representative" ], "overview_url": "https://learn.microsoft.com/en-us/dynamics365/contact-center/administer/set-up-voice-agents-to-use-intents", "configure_url": "https://learn.microsoft.com/en-us/dynamics365/contact-center/administer/set-up-voice-agents-to-use-intents", "doc_status": "dedicated", "source_slide": 2, "_catalog_kind": "first_party", "_documented": true, "_availability_label": "Available" }, { "id": "customer-knowledge-management-agent", "name": "Customer Knowledge Management Agent", "deck_label": "Customer Knowledge Mgmt Agent", "product": "Dynamics 365 Customer Service", "vertical": "customer_service", "deck_status": "GA", "deck_description": "Extracts knowledge from closed cases, fills gaps", "description": "Creates knowledge articles from cases and related conversations, emails, and notes.", "use_cases": [ "Create draft knowledge articles from resolved cases and related communications", "Harvest knowledge from conversations, emails, and notes associated with a case", "Run knowledge harvesting in real time or through configured processing", "Review generated article content before it enters the governed knowledge lifecycle" ], "personas": [ "Knowledge Manager", "Service Manager", "Customer Service Representative" ], "overview_url": "https://learn.microsoft.com/en-us/dynamics365/customer-service/administer/admin-km-agent", "configure_url": "https://learn.microsoft.com/en-us/dynamics365/customer-service/administer/admin-km-agent", "doc_status": "dedicated", "source_slide": 2, "_catalog_kind": "first_party", "_documented": true, "_availability_label": "Available" }, { "id": "quality-assurance-agent", "name": "Quality Assurance Agent", "product": "Dynamics 365 Contact Center", "vertical": "customer_service", "deck_status": "Preview", "deck_description": "Real-time quality eval & coaching (CC umbrella)", "description": "Monitors live contact-center conversations, alerts supervisors to quality issues, and supports real-time coaching.", "use_cases": [ "Monitor active customer conversations for quality issues", "Alert supervisors when a quality indicator or guardrail needs attention", "Run real-time or conversation-close evaluation plans", "Send coaching nudges to representatives during active conversations" ], "personas": [ "Contact Center Supervisor", "Quality Manager", "Customer Service Representative" ], "overview_url": "https://learn.microsoft.com/en-us/dynamics365/contact-center/administer/use-quality-assurance-agent", "configure_url": "https://learn.microsoft.com/en-us/dynamics365/contact-center/administer/configure-quality-coach", "doc_status": "dedicated", "source_slide": 2, "_catalog_kind": "first_party", "_documented": true, "_availability_label": "Available" }, { "id": "quality-evaluation-agent", "name": "Quality Evaluation Agent", "product": "Dynamics 365 Contact Center / Customer Service", "vertical": "customer_service", "deck_status": "GA", "deck_description": "AI-scored assessments of interactions (cases + conversations)", "description": "Uses supervisor-defined criteria and evaluation plans to score selected interactions and provide summaries and recommended actions.", "use_cases": [ "Define reusable quality criteria and scoring logic", "Create evaluation plans that select which cases or conversations are assessed", "Generate evaluation scores, summaries, and recommended actions", "Review quality trends and outliers across evaluated interactions" ], "capability_notes": [ "Email evaluation is explicitly documented as Preview." ], "personas": [ "Service Supervisor", "Quality Analyst", "Contact Center Manager" ], "overview_url": "https://learn.microsoft.com/en-us/dynamics365/contact-center/administer/manage-quality-evaluation-agent", "configure_url": "https://learn.microsoft.com/en-us/dynamics365/contact-center/use/evaluation-plan", "doc_status": "dedicated", "source_slide": 2, "_catalog_kind": "first_party", "_documented": true, "_availability_label": "Available" }, { "id": "quality-management-agent", "name": "Quality Management Agent", "product": "Dynamics 365 Customer Service", "vertical": "customer_service", "deck_status": "Preview", "deck_description": "Governance + screen/call recording + eval (CS umbrella)", "description": "Provides a unified set of quality-management capabilities for evaluation, communication governance, and screen and call recording.", "use_cases": [ "Evaluate customer interactions against defined quality criteria", "Apply governance policies to outgoing customer communications", "Capture screen and call context for quality review", "Support coaching, compliance, and continuous service improvement" ], "personas": [ "Service Manager", "Quality Analyst", "Compliance Manager" ], "overview_url": "https://learn.microsoft.com/en-us/dynamics365/contact-center/use/overview-quality-management", "doc_status": "dedicated", "configure_note": "Quality Management Agent is configured through its individual quality evaluation, governance, and recording capabilities, which are linked from the overview.", "source_slide": 2, "_catalog_kind": "first_party", "_documented": true, "_availability_label": "Available" }, { "id": "account-reconciliation-agent", "name": "Account Reconciliation Agent", "product": "Dynamics 365 Finance", "vertical": "general", "deck_status": "Preview", "description": "Continuously compares general ledger transactions against subsidiary ledgers, flags discrepancies, and recommends corrective actions to keep the books reconciled between close periods.", "use_cases": [ "Continuously compare general ledger and subsidiary ledger transactions across the accounting period", "Flag discrepancies such as unmatched entries or amount mismatches as soon as they occur", "Review AI-suggested corrective actions before applying them to a reconciliation exception", "See a dashboard of open and closed reconciliation exceptions instead of waiting for period-end" ], "personas": [ "Accountant", "Controller", "Finance Manager" ], "overview_url": "https://learn.microsoft.com/en-us/dynamics365/finance/general-ledger/acct-rec-agent", "doc_status": "dedicated", "source_slide": 3, "lifecycle": "Preview", "lifecycle_source_url": "https://learn.microsoft.com/en-us/dynamics365/finance/general-ledger/acct-rec-agent", "_catalog_kind": "first_party", "_documented": true, "_availability_label": "Preview" }, { "id": "approvals-agent", "name": "Approvals Agent", "product": "Dynamics 365 Project Operations", "vertical": "general", "deck_status": "Preview", "description": "Reviews submitted time, expense, and material entries against uploaded policy documents and classifies each as ready for approval or needing review.", "use_cases": [ "Upload policy documents at the organization or project level for the agent to check entries against", "Automatically classify incoming time, expense, and material entries as ready for approval or needing review", "Optionally let the agent auto-approve entries that clearly match policy", "Reduce the manual review load on project managers approving routine entries" ], "personas": [ "Project Manager", "Approver", "Project Operations Administrator" ], "overview_url": "https://learn.microsoft.com/en-us/dynamics365/project-operations/approvals/approvals-agent-intro", "configure_url": "https://learn.microsoft.com/en-us/dynamics365/project-operations/approvals/approvals-agent-policy", "doc_status": "dedicated", "source_slide": 3, "lifecycle": "Preview", "lifecycle_source_url": "https://learn.microsoft.com/en-us/dynamics365/project-operations/approvals/approvals-agent-intro", "_catalog_kind": "first_party", "_documented": true, "_availability_label": "Preview" }, { "id": "expense-agent-project-operations", "name": "Expense Agent", "product": "Dynamics 365 Project Operations", "vertical": "general", "deck_status": "Preview", "description": "Extracts data from submitted receipts and assembles expense lines and reports for approval, reducing manual expense entry for project teams.", "use_cases": [ "Submit a receipt by uploading it in the app or forwarding it to a monitored mailbox", "Let the agent extract merchant, amount, currency, and date from the receipt automatically", "Group extracted expenses into a report by project or trip for review", "Apply configured policy rules before an expense report reaches an approver" ], "personas": [ "Project Team Member", "Project Manager", "Finance Approver" ], "overview_url": "https://learn.microsoft.com/en-us/dynamics365/project-operations/expense/expense-agent-overview", "configure_url": "https://learn.microsoft.com/en-us/dynamics365/project-operations/expense/expense-agent-configuration", "doc_status": "dedicated", "source_slide": 3, "lifecycle": "Preview", "lifecycle_source_url": "https://learn.microsoft.com/en-us/dynamics365/project-operations/expense/expense-agent-overview", "_catalog_kind": "first_party", "_documented": true, "_availability_label": "Preview" }, { "id": "expense-agent-business-central", "name": "Expense Agent", "product": "Dynamics 365 Business Central", "vertical": "general", "deck_status": "Preview", "description": "Lets any employee submit a receipt by web app or email, then extracts, categorizes, and assembles the data into an expense report inside Business Central.", "use_cases": [ "Submit a receipt through the dedicated expense web app or by forwarding it to a shared mailbox, with no Business Central license required", "Let the agent extract and categorize receipt data automatically", "Review and submit the resulting expense report for approval", "Keep all financial data, audit trails, and posting logic inside Business Central's existing rules" ], "personas": [ "Employee", "Expense Approver", "Finance Team" ], "overview_url": "https://learn.microsoft.com/en-us/dynamics365/business-central/expense-management/expense-agent", "doc_status": "dedicated", "source_slide": 3, "lifecycle": "Preview", "lifecycle_source_url": "https://learn.microsoft.com/en-us/dynamics365/business-central/expense-management/expense-agent", "_catalog_kind": "first_party", "_documented": true, "_availability_label": "Preview" }, { "id": "payables-agent", "name": "Payables Agent", "product": "Dynamics 365 Business Central", "vertical": "general", "deck_status": "Preview", "description": "Monitors a mailbox for vendor invoices, extracts the data with AI, and prepares a draft purchase invoice for a supervisor to review before posting.", "use_cases": [ "Monitor a designated mailbox for incoming vendor invoice emails with PDF attachments", "Extract invoice details automatically using AI-based document analysis", "Match the invoice to an existing vendor or flag it for supervisor input", "Review the draft purchase invoice before it is finalized and posted" ], "personas": [ "Accounts Payable Clerk", "Agent Supervisor", "Finance Approver" ], "overview_url": "https://learn.microsoft.com/en-us/dynamics365/business-central/payables-agent", "configure_url": "https://learn.microsoft.com/en-us/dynamics365/business-central/payables-agent-setup", "doc_status": "dedicated", "source_slide": 3, "lifecycle": "Preview", "lifecycle_source_url": "https://learn.microsoft.com/en-us/dynamics365/business-central/payables-agent", "_catalog_kind": "first_party", "_documented": true, "_availability_label": "Preview" }, { "id": "procurement-agent", "name": "Procurement Agent", "product": "Dynamics 365 Supply Chain Management", "vertical": "general", "deck_status": "Preview", "description": "Automates supplier email communications and evaluates the downstream impact of supplier-driven changes to purchase orders and supply commitments.", "use_cases": [ "Classify and analyze inbound supplier emails about purchase order confirmations and change requests", "Draft or send outbound reminder and follow-up emails to suppliers on the purchaser's behalf", "Evaluate how a supplier's requested change would affect inventory, production, and customer commitments", "Let purchasers quickly confirm and apply supplier-driven changes instead of re-entering them manually" ], "personas": [ "Purchasing Agent", "Procurement Manager", "Supply Chain Planner" ], "overview_url": "https://learn.microsoft.com/en-us/dynamics365/supply-chain/procurement/procurement-agent-overview", "doc_status": "dedicated", "source_slide": 3, "lifecycle": "Preview", "lifecycle_source_url": "https://learn.microsoft.com/en-us/dynamics365/supply-chain/procurement/procurement-agent-overview", "_catalog_kind": "first_party", "_documented": true, "_availability_label": "Preview" }, { "id": "sales-order-agent", "name": "Sales Order Agent", "product": "Dynamics 365 Business Central", "vertical": "general", "deck_status": "GA", "description": "Reads customer sales requests from a monitored mailbox, checks inventory, and prepares a sales quote or order for review before it's sent.", "use_cases": [ "Monitor a designated mailbox for incoming customer sales requests", "Clarify missing order details with the customer through a multi-turn email conversation", "Check inventory availability and prepare a sales quote or order as a PDF", "Let an authorized user review and approve the draft before it is converted to an order" ], "personas": [ "Sales Order Processor", "Customer Service Representative", "Sales Administrator" ], "overview_url": "https://learn.microsoft.com/en-us/dynamics365/business-central/sales-order-agent", "configure_url": "https://learn.microsoft.com/en-us/dynamics365/business-central/sales-order-agent-setup", "doc_status": "dedicated", "source_slide": 3, "_catalog_kind": "first_party", "_documented": true, "_availability_label": "Available" }, { "id": "time-entry-agent", "name": "Time Entry Agent", "product": "Dynamics 365 Project Operations", "vertical": "general", "deck_status": "Preview", "description": "Drafts weekly time entries for project team members from assignments, booked capacity, or prior entries, then reminds them to review and submit.", "use_cases": [ "Generate draft time entries at the start of each week from project assignments or booked capacity", "Notify team members in Microsoft Teams about their draft entries", "Review, modify, and submit draft entries from Teams or the Project Operations web app", "Receive a reminder near week's end if any time entries are still missing" ], "personas": [ "Project Team Member", "Resource Manager", "Project Operations Administrator" ], "overview_url": "https://learn.microsoft.com/en-us/dynamics365/project-operations/time/enable-time-entry-agent", "doc_status": "dedicated", "source_slide": 3, "lifecycle": "Preview", "lifecycle_source_url": "https://learn.microsoft.com/en-us/dynamics365/project-operations/time/enable-time-entry-agent", "_catalog_kind": "first_party", "_documented": true, "_availability_label": "Preview" } ], "partners": [ { "id": "congruentx", "name": "CongruentX", "url": "https://congruentx.com/ai-agents-library/", "description": "Microsoft Dynamics 365 and Power Platform partner publishing a public library of 40+ Copilot-based sales, insurance, distribution, and professional-services agents." } ], "partner_agents": [ { "id": "congruentx-buyer-intent-enrichment", "partner_id": "congruentx", "name": "Buyer Intent Enrichment (web + CRM)", "category": "b2b_sales", "product": "Copilot Studio \u00b7 Dynamics 365 Sales", "description": "Enriches anonymous web and CRM signals into scored buyer intent so marketing and sales can prioritize outreach to the accounts showing real interest.", "source_url": "https://congruentx.com/ai-agents-library/", "what_it_does": [ "Detects buying intent from web and CRM activity", "Prioritizes outreach toward the accounts showing the strongest intent", "Personalizes content and timing to match where each account is in its journey" ], "problem_it_solves": [ "Anonymous intent signals go uncaptured", "Lead quality and volume are inconsistent", "Generic content underperforms with in-market buyers" ], "who_it_helps": [ "Sales Rep, Manager", "Marketing", "Customer Success", "Ops" ], "partner_reported_outcome": "20\u201350% higher campaign ROI vs broad programs", "_catalog_kind": "partner" }, { "id": "congruentx-dynamic-journey-orchestration", "partner_id": "congruentx", "name": "Dynamic Journey Orchestration (CI Journeys Copilot)", "category": "b2b_sales", "product": "Copilot Studio \u00b7 Dynamics 365 Marketing", "description": "Orchestrates multi-channel nurture journeys and next-best-action sequencing across email, web, and ads based on real-time engagement signals.", "source_url": "https://congruentx.com/ai-agents-library/", "what_it_does": [ "Surfaces ICP fit and segment insights", "Aggregates research to equip early discovery conversations", "Builds next-best-action sequences across email, web, and ads", "Auto-enrolls segments into targeted journeys" ], "problem_it_solves": [ "ICP and message-market fit are fragmented across channels", "Discovery quality is low and leads are poorly qualified", "Manual journey-building slows down campaign cycles" ], "who_it_helps": [ "Sales Rep, Manager", "Marketing", "Customer Success", "Ops" ], "partner_reported_outcome": "10\u201315% lift in journey revenue", "_catalog_kind": "partner" }, { "id": "congruentx-event-webinar-persona-journeys", "partner_id": "congruentx", "name": "Event/Webinar Persona Journeys", "category": "b2b_sales", "product": "Copilot Studio \u00b7 Dynamics 365 Marketing", "description": "Builds automated, persona-specific follow-up journeys for event and webinar attendees based on how engaged each attendee was.", "source_url": "https://congruentx.com/ai-agents-library/", "what_it_does": [ "Builds automated follow-up journeys segmented by persona and engagement level" ], "problem_it_solves": [ "Event and webinar follow-up is created manually and inconsistently" ], "who_it_helps": [ "Sales Rep, Manager", "Marketing", "Customer Success" ], "partner_reported_outcome": "+10\u201320% higher MQL\u2192SQL conversion on event leads", "_catalog_kind": "partner" }, { "id": "congruentx-hyper-targeted-market-segmentation", "partner_id": "congruentx", "name": "Hyper-Targeted Market Segmentation", "category": "b2b_sales", "product": "Copilot Studio \u00b7 Dynamics 365 Marketing", "description": "Segments the addressable market at a granular level so campaigns can be targeted to the highest-propensity segments instead of broad, unfocused programs.", "source_url": "https://congruentx.com/ai-agents-library/", "what_it_does": [ "Detects intent and prioritizes outreach by market segment", "Personalizes content and timing to each segment" ], "problem_it_solves": [ "Intent signals go uncaptured at the segment level", "Marketing spend is spread across low-fit segments", "Generic content underperforms across a broad audience" ], "who_it_helps": [ "Sales Rep, Manager", "Marketing", "Customer Success", "Ops" ], "partner_reported_outcome": "+5\u201315% marketing ROI, reduced spend on low-impact segments", "_catalog_kind": "partner" }, { "id": "congruentx-first-impact-claims-portal", "partner_id": "congruentx", "name": "First-Impact: Claims Setup & Portal", "category": "financial_services", "product": "Copilot Studio \u00b7 Dynamics 365 Customer Service", "description": "Guides a policyholder through their first claim and first portal use so the initial experience builds trust instead of confusion.", "source_url": "https://congruentx.com/ai-agents-library/", "what_it_does": [ "Guides first claim and first portal use step by step" ], "problem_it_solves": [ "First-time claim and portal experiences lack visibility and guidance", "Engagement and volume are inconsistent during early claim interactions" ], "who_it_helps": [ "Sales Rep, Manager", "Marketing", "Customer Success", "Ops" ], "partner_reported_outcome": "+5\u201310 NPS points post-claim, better retention", "_catalog_kind": "partner" }, { "id": "congruentx-icp-targeting-by-line", "partner_id": "congruentx", "name": "ICP Targeting by Line (P&C/Life)", "category": "financial_services", "product": "Copilot Studio \u00b7 Dynamics 365 Sales", "description": "Targets prospects by ideal-customer-profile fit for a specific insurance line (P&C or Life), aggregating research to equip early discovery conversations.", "source_url": "https://congruentx.com/ai-agents-library/", "what_it_does": [ "Surfaces ICP fit and segment insights by insurance line", "Aggregates research to equip early discovery conversations" ], "problem_it_solves": [ "ICP and message-market fit are fragmented across lines of business", "Discovery quality is low and leads are poorly qualified", "Manual research slows down the sales cycle" ], "who_it_helps": [ "Sales Rep, Manager", "Marketing", "Customer Success", "Ops", "Underwriter / Claims" ], "partner_reported_outcome": "10\u201315% higher win rate", "_catalog_kind": "partner" }, { "id": "congruentx-risk-loss-ratio-storytelling", "partner_id": "congruentx", "name": "Risk & Loss Ratio Storytelling", "category": "financial_services", "product": "Copilot Studio \u00b7 Dynamics 365 Customer Service", "description": "Turns loss-ratio and risk data into a next-best-action narrative so underwriters and account teams can explain risk trends and engage the right stakeholders.", "source_url": "https://congruentx.com/ai-agents-library/", "what_it_does": [ "Surfaces ICP fit and segment insights from loss-ratio data", "Aggregates research to equip risk conversations", "Builds next-best-action sequences across channels" ], "problem_it_solves": [ "Loss-ratio and risk data is fragmented across systems", "Risk conversations lack supporting research", "Manual analysis slows down underwriting and account reviews" ], "who_it_helps": [ "Sales Rep, Manager", "Marketing", "Customer Success", "Ops" ], "_catalog_kind": "partner" }, { "id": "congruentx-target-accounts-buying-committee", "partner_id": "congruentx", "name": "Target Accounts by Buying Committee", "category": "professional_services", "product": "Copilot Studio \u00b7 Dynamics 365 Project Operations", "description": "Maps the buying committee inside a target account and prioritizes outreach to the specific roles involved in the decision.", "source_url": "https://congruentx.com/ai-agents-library/", "what_it_does": [ "Detects buying-committee membership and prioritizes outreach", "Personalizes content and timing per stakeholder role" ], "problem_it_solves": [ "Buying-committee signals go uncaptured", "Lead quality and volume are inconsistent", "Generic content underperforms with mixed-role buying committees" ], "who_it_helps": [ "Sales Rep, Manager", "Marketing", "Customer Success", "Ops", "Engagement Manager / PM", "Practice Lead" ], "partner_reported_outcome": "Higher win rate, larger average deal size (5\u201315%)", "_catalog_kind": "partner" }, { "id": "congruentx-thought-leadership-orchestration", "partner_id": "congruentx", "name": "Thought Leadership Orchestration", "category": "professional_services", "product": "Copilot Studio \u00b7 Dynamics 365 Marketing", "description": "Orchestrates thought-leadership content and next-best-action sequencing across channels to build pipeline influence over a longer sales cycle.", "source_url": "https://congruentx.com/ai-agents-library/", "what_it_does": [ "Surfaces ICP fit and segment insights for content targeting", "Aggregates research to equip early discovery", "Builds next-best-action sequences across email, web, and ads", "Auto-enrolls segments into targeted content journeys" ], "problem_it_solves": [ "ICP and message-market fit are fragmented across content channels", "Discovery quality is low without supporting research", "Manual content sequencing slows down pipeline influence" ], "who_it_helps": [ "Sales Rep, Manager", "Marketing", "Customer Success", "Ops", "Engagement Manager / PM", "Practice Lead" ], "partner_reported_outcome": "Strong influence on pipeline; higher engagement, 5\u201310% more opps sourced", "_catalog_kind": "partner" }, { "id": "congruentx-dormant-account-reactivation", "partner_id": "congruentx", "name": "Dormant Account Reactivation (Leads)", "category": "retail_cpg", "product": "Copilot Studio \u00b7 Dynamics 365 Marketing", "description": "Finds inactive accounts and builds a re-engagement play with an updated offer and story to win them back.", "source_url": "https://congruentx.com/ai-agents-library/", "what_it_does": [ "Finds and re-engages inactive accounts", "Builds a targeted, updated offer and story for reactivation" ], "problem_it_solves": [ "Dormant accounts go untouched without a systematic reactivation play", "Offers and messaging go stale for lapsed accounts" ], "who_it_helps": [ "Sales Rep, Manager", "Marketing", "Customer Success", "Ops" ], "partner_reported_outcome": "10\u201320% of dormant accounts reactivated", "_catalog_kind": "partner" }, { "id": "congruentx-loyalty-tier-triggers", "partner_id": "congruentx", "name": "Loyalty Tier Triggers \u2192 Offers", "category": "retail_cpg", "product": "Copilot Studio \u00b7 Dynamics 365 Customer Service", "description": "Watches loyalty-tier signals and automatically triggers the next relevant offer or advocacy ask at the right moment in the customer relationship.", "source_url": "https://congruentx.com/ai-agents-library/", "what_it_does": [ "Surfaces cross-sell and upsell moments from loyalty-tier signals", "Activates advocacy and reference requests at the right moment" ], "problem_it_solves": [ "No systematic trigger exists for cross-sell moments", "Advocates and references are hard to find", "Attach rates and loyalty stay flat without a structured trigger" ], "who_it_helps": [ "Sales Rep, Manager", "Marketing", "Customer Success", "Ops", "Branch / Inside Sales", "Supply Chain" ], "_catalog_kind": "partner" }, { "id": "congruentx-service-contracts-attach-rate", "partner_id": "congruentx", "name": "Service Contracts & Attach Rate Signals", "category": "retail_cpg", "product": "Copilot Studio \u00b7 Dynamics 365 Sales", "description": "Prompts the right warranty or service contract at the point of sale, using signal detection to raise attach rates without a manual upsell script.", "source_url": "https://congruentx.com/ai-agents-library/", "what_it_does": [ "Prompts warranties and service contracts at the point of sale" ], "problem_it_solves": [ "Visibility into attach-rate benchmarks is limited", "Retention stays low without a point-of-sale service prompt" ], "who_it_helps": [ "Sales Rep, Manager", "Marketing", "Customer Success", "Ops" ], "partner_reported_outcome": "10\u201330% higher attach rate", "_catalog_kind": "partner" } ], "community_tools": [ { "id": "business-process-catalog", "name": "Business Process Catalog", "url": "https://learn.microsoft.com/en-us/dynamics365/guidance/business-processes/about", "author": "Microsoft", "description": "Official Microsoft Dynamics 365 reference catalog of standard business processes, importable into Azure DevOps, useful for mapping an implementation's process scope before deciding what to automate with an agent." }, { "id": "copilot-ramp-playbook", "name": "Copilot Ramp Playbook", "url": "https://github.com/mfm-copilot-ramp/copilot-ramp-playbook", "author": "mfm-copilot-ramp", "description": "Community learning guide covering seven stages from Copilot chat through agent development, with role-based paths, solution examples, and cost/value planning tools. This is not an official Microsoft resource." }, { "id": "copilot-streaming-chat-playground", "name": "Copilot Streaming Chat Playground", "url": "https://github.com/jzh24516/copilot-streaming-chat-playground", "author": "jzh24516", "description": "Community playground project for experimenting with streaming chat responses against Copilot-style APIs." }, { "id": "cowork-cookbook-prospect-to-quote", "name": "Cowork Cookbook \u2014 Prospect to Quote", "url": "https://coworkcookbook.com/map/prospect-to-quote", "author": "Cowork Cookbook", "description": "Community recipe map walking through a prospect-to-quote workflow, useful as a reference pattern for structuring a similar sales agent flow." } ] }