{
  "schema": "aibast-canonical-transcripts/1.0",
  "captured_at": "2026-10-07T01:24:35.566942+00:00",
  "solution": "@aibast-agents-library/sales-qualification",
  "onepager": "21. Sales Qualification Agent one-pager.pptx",
  "brainstem_version": "0.6.16",
  "loaded_tools_after_capture": [
    "SalesQualificationAgent"
  ],
  "agent_sources": [
    {
      "path": "agents/@aibast-agents-library/b2b_sales_stacks/sales_qualification_stack/sales_qualification_agent.py",
      "sha256": "5ec782d1e5ef9906de2e37a624c238933408b3fa866ffa25e5194eb46ad75141"
    }
  ],
  "case_file": "tests/demo_cases/sales-qualification.json",
  "case_file_sha256": "0b74cd9adaaa2fa264e5312ea66d751b690d536f6b5d24c2c543968a8606eca3",
  "strict_isolation": true,
  "transcripts": [
    {
      "case_id": "SQ-01",
      "persona": "Sales Manager",
      "onepager_promise": "Perform intelligent lead enrichment to prioritize high-potential opportunities",
      "prompt": "Which bundled synthetic leads should my team review first, and why did they score that way?",
      "assistant_response": "Your team should start with the **8 Hot leads (score 80+)**, especially the top 3:\n\n- **TechFlow Industries \u2014 94**: VP of Engineering contact, actively evaluating solutions now, strong tech-stack fit (AWS/Snowflake/Kubernetes), highest intent signal (98).\n- **Meridian Corp \u2014 91**: CTO-level contact with budget already approved \u2014 a strong buying-stage signal, large org (1,200 employees).\n- **Apex Solutions \u2014 88**: Competitor-displacement opportunity (actively looking to switch), high intent (98), solid stack fit.\n\nScoring blends **35% ICP fit + 25% BANT + 40% intent data**, so these rose to the top because they combine the right company profile, real buying signals (budget/authority), and high recent engagement \u2014 not just firmographic match alone.\n\nBelow them: 15 Warm leads (60-79) worth an SDR call, and 22 Nurture leads better suited to an email sequence for now.\n\nWant me to run the BANT breakdown on these top hot leads next, or move straight to drafting outreach?",
      "agent_logs": "[SalesQualificationAgent] **Lead Qualification Summary \u2014 45 Leads Scored**\n\nAnalyzed 45 leads with ICP scoring and BANT criteria plus intent data (score = 35% ICP fit + 25% BANT + 40% intent).\n\n| Tier | Leads | Recommended Action |\n|---|---|---|\n| Hot (80+) | 8 | AE handoff |\n| Warm (60-79) | 15 | SDR call |\n| Nurture (<60) | 22 | Email sequence |\n\n**Top 3 Hot Leads:**\n- **TechFlow Industries** (94) - VP Eng, active eval\n- **Meridian Corp** (91) - CTO, budget approved\n- **Apex Solutions** (88) - Competitor displacement\n\n**Enrichment (firmographic, technographic, intent):**\n\n| Company | Employees | Industry | Tech Stack | Intent |\n|---|---|---|---|---|\n| TechFlow Industries | 520 | Technology | AWS, Snowflake, Kubernetes | 98 |\n| Meridian Corp | 1,200 | Healthcare | Azure, Salesforce, Databricks | 88 |\n| Apex Solutions | 780 | SaaS | AWS, Kubernetes, Salesforce | 98 |\n\nNext: BANT analysis on the hot leads.\n\nSynthetic source model: [CRM + ZoomInfo + 6sense Intent Data]\nAgents: LeadEnrichmentAgent, ICPMatchingAgent\n\n**Evidence boundary:** Exact names, lead counts, company attributes, scores, values, percentages, and timing are synthetic planning evidence. Outreach and routing are drafts for human review. No lead was assigned, no sequence or alert was activated, and no CRM or customer communication occurred.",
      "expected_agent": "SalesQualificationAgent",
      "must_include": [
        "Lead Qualification Summary",
        "Top 3 Hot Leads",
        "Hot (80+)",
        "Evidence boundary"
      ],
      "model": "claude-sonnet-5",
      "requested_model": "claude-sonnet-5",
      "passed": true
    },
    {
      "case_id": "SQ-02",
      "persona": "Business Development Rep.",
      "onepager_promise": "Automate qualification scoring to identify conversion-ready prospects",
      "prompt": "Show the BANT evidence and missing qualification details for the strongest synthetic leads.",
      "assistant_response": "Here's the BANT breakdown for the top 5 hot leads:\n\n| Lead | Budget | Authority | Need | Timeline | Score |\n|---|---|---|---|---|---|\n| TechFlow Industries | $200K | VP Eng | Consolidate 12 data sources | Q1 | 89 |\n| Meridian Corp | $150K | CTO | Replace legacy EHR layer | 60 days | 89 |\n| Apex Solutions | $180K | Director | Displace incumbent vendor | Q1 | 76 |\n| DataCorp Analytics | $90K | IT Manager | +40% pipeline efficiency | Q2 | 69 |\n| Summit Technologies | TBD (~$40K) | VP Ops | Scale monitoring, 8 plants | 60 days | 62 |\n\n**Strongest signals:** TechFlow (repeat booth visits), Meridian (CTO asking technical questions), Apex (competitor contract expiring).\n\n**Missing/at-risk qualification details:**\n- **DataCorp Analytics** \u2014 contact is an IT Manager, not a confirmed decision-maker; needs authority verification.\n- **Summit Technologies** \u2014 budget is unconfirmed/estimated, needs discovery to lock in number.\n\nEverything here is draft synthetic data for your review \u2014 no CRM writes, outreach, or assignments have happened. Want me to generate outreach sequences or assign these to AEs next?",
      "agent_logs": "[SalesQualificationAgent] **BANT Analysis \u2014 Top 5 Hot Leads**\n\n| Lead | Budget | Authority | Need | Timeline | BANT Score |\n|---|---|---|---|---|---|\n| TechFlow Industries | $200K | VP Eng | Consolidate 12 data sources into unified pipe | Q1 | 89 |\n| Meridian Corp | $150K | CTO | Replace legacy EHR integration layer | 60 days | 89 |\n| Apex Solutions | $180K | Director | Displace incumbent vendor, contract ending Q1 | Q1 | 76 |\n| DataCorp Analytics | $90K | IT Manager | Improve data pipeline efficiency by 40% | Q2 | 69 |\n| Summit Technologies | TBD (est. $40K) | VP Operations | Scale production monitoring across 8 plants | 60 days | 62 |\n\n**Strongest Engagement Signals:**\n- **TechFlow Industries**: Demo booth visited twice\n- **Meridian Corp**: CTO asked technical questions\n- **Apex Solutions**: Competitor contract ending\n\n**Risk Flags:**\n- DataCorp Analytics: Needs a decision maker (IT Manager)\n- Summit Technologies: Budget TBD\n\nSynthetic source model: [CRM + Booth Interactions + Intent Data]\nAgents: BANTScoringAgent\n\n**Evidence boundary:** Exact names, lead counts, company attributes, scores, values, percentages, and timing are synthetic planning evidence. Outreach and routing are drafts for human review. No lead was assigned, no sequence or alert was activated, and no CRM or customer communication occurred.",
      "expected_agent": "SalesQualificationAgent",
      "must_include": [
        "BANT Analysis",
        "Strongest Engagement Signals",
        "$200K",
        "Evidence boundary"
      ],
      "model": "claude-sonnet-5",
      "requested_model": "claude-sonnet-5",
      "passed": true
    },
    {
      "case_id": "SQ-03",
      "persona": "Business Development Rep.",
      "onepager_promise": "Generate personalized outreach sequences to boost engagement",
      "prompt": "Draft outreach ideas for the synthetic hot leads, but do not send or schedule any communication.",
      "assistant_response": "Here's the draft outreach plan for the 8 hot leads \u2014 all read-only, nothing sent or scheduled:\n\n**Each lead gets a personalized hook + CTA**, e.g.:\n- TechFlow Industries (Sarah Nguyen, VP Eng) \u2192 \"Connecting 12 data sources in weeks\" \u2192 15-min deep dive\n- Meridian Corp (James Walker, CTO) \u2192 migration playbook hook \u2192 stack discussion\n- Apex, DataCorp, Summit, Greenfield, Orion, FusionTech \u2014 each with its own competitive/use-case hook and matching CTA\n\n**Proposed cadence (draft only):**\n1. Day 0 \u2013 Personalized email\n2. Day 2 \u2013 LinkedIn connect + note\n3. Day 3 \u2013 Value content email\n4. Day 4 \u2013 Phone call\n\nAll contact names, hooks, and timing are synthetic demo data for review \u2014 no emails, calls, or LinkedIn messages were actually sent or scheduled, and nothing was written to a CRM.\n\nWant me to move to AE assignment next?",
      "agent_logs": "[SalesQualificationAgent] **Personalized Outreach \u2014 8 Hot Leads**\n\nPersonalized outreach drafted for all 8 hot leads.\n\n| Lead | Contact | Personalized Hook | CTA |\n|---|---|---|---|\n| TechFlow Industries | Sarah Nguyen, VP Engineering | \"Connecting 12 data sources in weeks\" | 15-min deep dive CTA |\n| Meridian Corp | James Walker, CTO | \"60-day migration playbook attached\" | Stack discussion CTA |\n| Apex Solutions | Diana Reyes, Director of IT | \"40% of [Competitor] customers switched\" | Comparison call CTA |\n| DataCorp Analytics | Emily Tran, IT Manager | \"40% faster pipelines on your trial data\" | Trial review call CTA |\n| Summit Technologies | Robert Kim, VP Operations | \"Monitoring 8 plants from one pipeline\" | Plant-rollout walkthrough CTA |\n| Greenfield Health | Maria Santos, Chief Digital Officer | \"One patient record across 14 facilities\" | Architecture session CTA |\n| Orion Manufacturing | Thomas Park, CTO | \"Predictive maintenance from IoT data in 90 days\" | Pilot scoping call CTA |\n| FusionTech Labs | Derek Johnson, CTO | \"Hadoop to cloud-native migration assessment\" | Assessment kickoff CTA |\n\n**Draft Sequence Cadence (not activated):** Personalized email today > LinkedIn connection + note day 2 > Value content email day 3 > Phone call day 4\n- Today: Personalized email\n- Day 2: LinkedIn connection + note\n- Day 3: Value content email\n- Day 4: Phone call\n\nNext: assign the leads to AEs.\n\nSynthetic source model: [Content Library + Booth Notes + LinkedIn]\nAgents: PersonalizedOutreachAgent\n\n**Evidence boundary:** Exact names, lead counts, company attributes, scores, values, percentages, and timing are synthetic planning evidence. Outreach and routing are drafts for human review. No lead was assigned, no sequence or alert was activated, and no CRM or customer communication occurred.",
      "expected_agent": "SalesQualificationAgent",
      "must_include": [
        "Personalized Outreach",
        "Draft Sequence Cadence",
        "Connecting 12 data sources in weeks",
        "Evidence boundary"
      ],
      "model": "claude-sonnet-5",
      "requested_model": "claude-sonnet-5",
      "passed": true
    },
    {
      "case_id": "SQ-04",
      "persona": "Sales Manager",
      "onepager_promise": "Coordinate SLA-based assignments and tracking to drive timely engagement",
      "prompt": "Recommend synthetic lead routing for manager review without assigning CRM owners.",
      "assistant_response": "Here's the routing recommendation for your 8 hot leads \u2014 all drafted for manager review, nothing has been assigned yet:\n\n- **Mike Rodriguez** (Enterprise, 62% capacity) \u2014 TechFlow Industries ($200K), Apex Solutions ($180K), DataCorp Analytics ($90K)\n- **Sarah Kim** (Healthcare, 55% capacity) \u2014 Meridian Corp ($150K), Greenfield Health ($60K)\n- **James Chen** (Manufacturing, 70% capacity) \u2014 Summit Technologies ($40K), Orion Manufacturing ($60K)\n- **Lisa Park** (Mid-Market SaaS, 48% capacity) \u2014 FusionTech Labs ($20K)\n\nAll four AEs are under 80% capacity, matched by territory/industry expertise. Each gets a handoff package (BANT summary, booth notes, draft email).\n\nThis is a recommendation only \u2014 no leads are actually assigned in the CRM. Want me to set up SLA follow-up tracking next, or pull the full qualification report?",
      "agent_logs": "[SalesQualificationAgent] **Recommended Lead Routing \u2014 8 Hot Leads to 4 AEs**\n\nLeads routed by territory and expertise:\n\n| AE | Leads | Pipeline | Specialty | Capacity |\n|---|---|---|---|---|\n| Mike Rodriguez | 3 | $470K | Enterprise | 62% |\n| Sarah Kim | 2 | $210K | Healthcare | 55% |\n| James Chen | 2 | $100K | Manufacturing | 70% |\n| Lisa Park | 1 | $20K | Mid-Market SaaS | 48% |\n\n**Assignment Detail:**\n- Mike Rodriguez: TechFlow Industries ($200K), Apex Solutions ($180K), DataCorp Analytics ($90K)\n- Sarah Kim: Meridian Corp ($150K), Greenfield Health ($60K)\n- James Chen: Summit Technologies ($40K), Orion Manufacturing ($60K)\n- Lisa Park: FusionTech Labs ($20K)\n\nAll AEs under 80% capacity. Handoff packages include BANT summary, booth notes, and email drafts.\n\nNext: set up SLA tracking.\n\nSynthetic source model: [Territory Rules + Capacity Dashboard]\nAgents: LeadRoutingAgent\n\n**Evidence boundary:** Exact names, lead counts, company attributes, scores, values, percentages, and timing are synthetic planning evidence. Outreach and routing are drafts for human review. No lead was assigned, no sequence or alert was activated, and no CRM or customer communication occurred.",
      "expected_agent": "SalesQualificationAgent",
      "must_include": [
        "Recommended Lead Routing",
        "Handoff Package",
        "$470K",
        "Evidence boundary"
      ],
      "model": "claude-sonnet-5",
      "requested_model": "claude-sonnet-5",
      "passed": true
    },
    {
      "case_id": "SQ-05",
      "persona": "Sales Manager",
      "onepager_promise": "Coordinate SLA-based assignments and tracking to drive timely engagement",
      "prompt": "Draft an SLA and escalation plan for the synthetic leads without activating alerts or automations.",
      "assistant_response": "Here's the draft SLA and escalation plan \u2014 all read-only, nothing activated:\n\n**Response SLAs by tier (45 leads)**\n- **Hot (8 leads):** 4-hour response SLA \u2192 escalates to manager alert if missed\n- **Warm (15 leads):** 24-hour response SLA \u2192 escalates to team alert if missed\n- **Nurture (22 leads):** 48-hour response SLA \u2192 routes to auto-sequence if missed\n\n**Escalation triggers (proposed, not live)**\n- Alert at 2 hours remaining before SLA breach\n- Manager notification on any missed SLA\n- Suggested CRM stage update when a meeting gets booked\n\n**Success targets**\n- 100% hot-lead contact rate\n- 40% meeting conversion\n- $800K pipeline influenced\n\nNothing here is switched on \u2014 no alerts fire, no CRM gets written to, no automations run until you explicitly approve activation. Want me to pull together the full qualification report next, tying this SLA plan to the scoring, BANT, outreach, and AE assignment into one action plan?",
      "agent_logs": "[SalesQualificationAgent] **Draft SLA Tracking Plan \u2014 45 Synthetic Leads (ready to activate)**\n\n| Tier | Leads | Response SLA | Escalation |\n|---|---|---|---|\n| Hot | 8 | 4 hours | Manager alert |\n| Warm | 15 | 24 hours | Team alert |\n| Nurture | 22 | 48 hours | Auto-sequence |\n\n**Automations to switch on (not activated):** Teams alerts at 2 hr remaining, manager notification if an SLA is missed, CRM stage update suggested when a meeting is booked.\n\n**Targets:** 100% hot contact rate, 40% meeting conversion, $800K pipeline\n\nNext: generate the summary.\n\nSynthetic source model: [SLA Engine + Notification System]\nAgents: SLAMonitoringAgent\n\n**Evidence boundary:** Exact names, lead counts, company attributes, scores, values, percentages, and timing are synthetic planning evidence. Outreach and routing are drafts for human review. No lead was assigned, no sequence or alert was activated, and no CRM or customer communication occurred.",
      "expected_agent": "SalesQualificationAgent",
      "must_include": [
        "Draft SLA Tracking Plan",
        "Targets",
        "$800K pipeline",
        "Evidence boundary"
      ],
      "model": "claude-sonnet-5",
      "requested_model": "claude-sonnet-5",
      "passed": true
    },
    {
      "case_id": "SQ-06",
      "persona": "Account Executive",
      "onepager_promise": "Automate qualification scoring to identify conversion-ready prospects",
      "prompt": "Summarize the synthetic qualified pipeline and clearly label every conversion and value assumption.",
      "assistant_response": "Here's the synthetic pipeline summary \u2014 every figure below is bundled demo data, not real leads:\n\n**Pipeline breakdown (45 synthetic leads):**\n- \ud83d\udd25 **Hot: 8 leads = $800K** \u2014 labeled as pipeline value assumption, draft outreach prepared, no sends\n- \ud83c\udf24\ufe0f **Warm: 15 leads = $450K** \u2014 assumption-based value, slated for SDR calls\n- \ud83c\udf31 **Nurture: 22 leads** \u2014 no value assigned, routed to email sequences only\n\n**Total assumed pipeline: $1.25M** (Hot + Warm only \u2014 this is a projected/assumed figure, not closed or forecasted revenue)\n\n**Conversion assumptions (labeled, not guaranteed):**\n- Hot leads \u2192 AE outreach within 4-hour SLA (assumption: faster follow-up = higher close rate)\n- Warm leads \u2192 SDR call cadence (assumption: needs more qualification before AE handoff)\n- Nurture leads \u2192 automated email only (assumption: lower near-term conversion probability)\n\n**AE routing:** 4 reps recommended by territory/expertise \u2014 draft assignment, nothing written to CRM.\n\n**Everything here is synthetic, read-only, and awaiting your approval** \u2014 no outreach sent, no leads assigned, no CRM writes. Want me to move forward with AE assignment or activate the SLA tracking plan?",
      "agent_logs": "[SalesQualificationAgent] **Qualification Report \u2014 Lead qualification complete**\n\n| Result | Value |\n|---|---|\n| Leads analyzed | 45 |\n| Hot leads | 8 ($800K) |\n| Outreach drafted | All 8 hot leads |\n| AEs recommended | 4 reps |\n| SLA tracking | Plan ready to activate |\n\n**Total pipeline:** $1.25M (Hot $800K + Warm $450K)\n\n**Action plan:** 8 hot leads get AE outreach within 4 hours, 15 warm get SDR calls, 22 nurture enter email sequences.\n\n**Draft Review Queue:** approve the hot-lead outreach drafts and AE routing, then activate the SLA plan.\n\nSynthetic source model: [All Qualification Systems]\nAgents: QualificationReportAgent (orchestrating all agents)\n\n**Evidence boundary:** Exact names, lead counts, company attributes, scores, values, percentages, and timing are synthetic planning evidence. Outreach and routing are drafts for human review. No lead was assigned, no sequence or alert was activated, and no CRM or customer communication occurred.",
      "expected_agent": "SalesQualificationAgent",
      "must_include": [
        "Qualification Report",
        "$1.25M",
        "Action plan",
        "Evidence boundary"
      ],
      "model": "claude-sonnet-5",
      "requested_model": "claude-sonnet-5",
      "passed": true
    }
  ]
}