Create a blank Copilot Studio agent
Step 1 of 32
Action
Create a blank Copilot Studio agentExpected resultA blank Copilot Studio agent is visible in the captured Draft workspace.
Manual mode · literal browser construction
Build Customer Loyalty and Rewards Agent manually.
No PAC CLI, YAML import, or plugin architect. Follow each action and its evidence boundary, compare only the reviewed references, and stop at Draft.
Synthetic disclosure: this is qualitative workflow evidence using packaged synthetic inputs. It is not a customer KPI or a live-system result.
Progress will remain available on this page, but it will not persist after you leave.
Found something inaccurate? Use Report an issue on that step. It opens a prefilled GitHub issue for review and does not submit automatically.
Build and verify
Name Customer Loyalty Manual
Step 2 of 32
Action
Name Customer Loyalty ManualCustomer Loyalty Manual
Expected resultThe page header shows the recorded manual build name: Customer Loyalty Manual.
Enter the reviewed global instructions
Step 3 of 32
Action
Enter the reviewed global instructions# Customer Loyalty and Rewards Agent — Manual Global Instructions Use only the uploaded anonymous synthetic loyalty records, safety rules, and operation skills. Treat balances, tiers, and catalog items as informational. The demo program is a fictional 450K-member retailer program: at-risk churn segments, the top at-risk members (Linda M., Kevin R., Sarah T.), win-back offer drafts for three segments, a launch-ready campaign plan with projected 14-day results, four program improvements, and a session summary. Campaigns are prepared for the user to launch; never say a campaign was sent or launched. Never contact or enroll a member, change points or tier, create an offer, issue or redeem a reward, refund funds, create an order, or complete a purchase. Lead with program evidence, distinguish analysis from eligibility, name the authorized workflow required for action, and state that no side effect occurred. <!-- locked-preview-anchors:start --> ## Skill routing map Route from the user's natural-language intent to the correct skill below. Do not narrate internal retrieval, tool selection, restrictions, or implementation mechanics; present only the user-facing result. - `CLR-01` uses skill `synthetic-loyalty-program-health`. - `CLR-02` uses skill `informational-points-summary`. - `CLR-03` uses skill `review-only-reward-options`. - `CLR-04` uses skill `loyalty-tier-structure-analysis`. - `CLR-05` uses skill `at-risk-churn-segments`. - `CLR-06` uses skill `top-at-risk-member-profiles`. - `CLR-07` uses skill `win-back-offer-drafts`. - `CLR-08` uses skill `campaign-launch-plan`. - `CLR-09` uses skill `loyalty-program-improvements`. - `CLR-10` uses skill `loyalty-session-summary`. These skill names above are the ONLY valid skill identifiers. Never invent, guess, or reference any other skill name. If the correct skill or its knowledge cannot be loaded after one retry in the same turn, say so honestly and stop. Do not answer using values you already know from these instructions or from general knowledge -- a response with no real citation is not acceptable output. <!-- locked-preview-anchors:end -->
Expected resultThe reviewed manual/GLOBAL-INSTRUCTIONS.md policy is visible or saved without unrecorded edits.
Save the instruction policy
Step 4 of 32
Action
Save the instruction policyExpected resultThe reviewed manual/GLOBAL-INSTRUCTIONS.md policy is visible or saved without unrecorded edits.
Remove default web search
Step 5 of 32
Action
Remove default web searchExpected resultThe captured inventory no longer lists the default web-search capability.
Add knowledge: aibast_loyalty-rewards-rules-and-safety.md
Step 6 of 32
Action
Add knowledge: aibast_loyalty-rewards-rules-and-safety.mdExpected resultThe exact knowledge file linked for this step is attached under its complete filename. Wait for processing to finish before continuing. This step does not require the final inventory or prove the other uploads.
Add knowledge: aibast_loyalty-rewards-synthetic-records.md
Step 7 of 32
Action
Add knowledge: aibast_loyalty-rewards-synthetic-records.mdExpected resultThe exact knowledge file linked for this step is attached under its complete filename. Wait for processing to finish before continuing. This step does not require the final inventory or prove the other uploads.
Add skill: campaign-plan
Step 8 of 32
Action
Add skill: campaign-planExpected resultThe linked SKILL.md appears with its frontmatter name and description. Open it and compare the complete instructions with the source file. This step does not require the final inventory or prove the other uploads.
Add skill: campaign-summary
Step 9 of 32
Action
Add skill: campaign-summaryExpected resultThe linked SKILL.md appears with its frontmatter name and description. Open it and compare the complete instructions with the source file. This step does not require the final inventory or prove the other uploads.
Add skill: churn-risk-segments
Step 10 of 32
Action
Add skill: churn-risk-segmentsExpected resultThe linked SKILL.md appears with its frontmatter name and description. Open it and compare the complete instructions with the source file. This step does not require the final inventory or prove the other uploads.
Add skill: loyalty-dashboard
Step 11 of 32
Action
Add skill: loyalty-dashboardExpected resultThe linked SKILL.md appears with its frontmatter name and description. Open it and compare the complete instructions with the source file. This step does not require the final inventory or prove the other uploads.
Add skill: points-summary
Step 12 of 32
Action
Add skill: points-summaryExpected resultThe linked SKILL.md appears with its frontmatter name and description. Open it and compare the complete instructions with the source file. This step does not require the final inventory or prove the other uploads.
Add skill: program-improvements
Step 13 of 32
Action
Add skill: program-improvementsExpected resultThe linked SKILL.md appears with its frontmatter name and description. Open it and compare the complete instructions with the source file. This step does not require the final inventory or prove the other uploads.
Add skill: reward-recommendations
Step 14 of 32
Action
Add skill: reward-recommendationsExpected resultThe linked SKILL.md appears with its frontmatter name and description. Open it and compare the complete instructions with the source file. This step does not require the final inventory or prove the other uploads.
Add skill: tier-analysis
Step 15 of 32
Action
Add skill: tier-analysisExpected resultThe linked SKILL.md appears with its frontmatter name and description. Open it and compare the complete instructions with the source file. This step does not require the final inventory or prove the other uploads.
Add skill: top-at-risk-members
Step 16 of 32
Action
Add skill: top-at-risk-membersExpected resultThe linked SKILL.md appears with its frontmatter name and description. Open it and compare the complete instructions with the source file. This step does not require the final inventory or prove the other uploads.
Add skill: winback-offers
Step 17 of 32
Action
Add skill: winback-offersExpected resultThe linked SKILL.md appears with its frontmatter name and description. Open it and compare the complete instructions with the source file. This step does not require the final inventory or prove the other uploads.
Review model, skills, knowledge, and safety boundaries
Step 18 of 32
Action
Review model, skills, knowledge, and safety boundariesExpected resultThe screenshot visibly confirms Claude Opus 5, 10 rendered skills, and no tools. Knowledge is verified separately in the two knowledge-upload checkpoints.
Review the skill list (10 skills)
Step 19 of 32
Action
Review the skill list (10 skills)Expected resultThe screenshot visibly confirms Claude Opus 5, 10 rendered skills, and no tools. Knowledge is verified separately in the two knowledge-upload checkpoints.
Review the rest of the skill list
Step 20 of 32
Action
Review the rest of the skill listExpected resultThe screenshot visibly confirms Claude Opus 5, 10 rendered skills, and no tools. Knowledge is verified separately in the two knowledge-upload checkpoints.
Open a fresh Preview conversation
Step 21 of 32
Action
Open a fresh Preview conversationExpected resultA fresh Preview surface or its recorded qualitative result is visible; only the evidence file defines a pass.
Pass CLR-01: Loyalty Program Director
Step 22 of 32
Action
Pass CLR-01: Loyalty Program DirectorAs Loyalty Program Director, summarize the synthetic program health and governance boundary.
Expected resultThe captured Preview evidence records CLR-01 with the recorded identifiers Prepared for:** Loyalty Program Director, Synthetic Loyalty Program Dashboard, no member is contacted; do not infer results beyond it.
Before this case
Open Preview in a fresh conversation before running CLR-01. A previous response must not influence the evidence.
Pass CLR-02: CRM Manager
Step 23 of 32
Action
Pass CLR-02: CRM ManagerAs CRM Manager, explain the synthetic Gold member balance without changing the ledger.
Expected resultThe captured Preview evidence records CLR-02 with the recorded identifiers Prepared for:** CRM Manager, Synthetic Points Summary, Synthetic Gold Member; do not infer results beyond it.
Before this case
Open Preview in a fresh conversation before running CLR-02. A previous response must not influence the evidence.
Pass CLR-03: Marketing Leader
Step 24 of 32
Action
Pass CLR-03: Marketing LeaderAs Marketing Leader, show review-only reward options without creating offers or issuing benefits.
Expected resultThe captured Preview evidence records CLR-03 with the recorded identifiers Prepared for:** Marketing Leader, Draft Reward Option Recommendations, no points, tier, offer, reward; do not infer results beyond it.
Before this case
Open Preview in a fresh conversation before running CLR-03. A previous response must not influence the evidence.
Pass CLR-04: Loyalty Program Director
Step 25 of 32
Action
Pass CLR-04: Loyalty Program DirectorAs Loyalty Program Director, review the synthetic tier structure without changing member status.
Expected resultThe captured Preview evidence records CLR-04 with the recorded identifiers Synthetic Tier Analysis, Tier Structure, no member is contacted; do not infer results beyond it.
Before this case
Open Preview in a fresh conversation before running CLR-04. A previous response must not influence the evidence.
Pass CLR-05: Marketing Leader
Step 26 of 32
Action
Pass CLR-05: Marketing LeaderAs Marketing Leader, help me identify loyalty members who are at risk of churning.
Expected resultThe captured Preview evidence records CLR-05 with the recorded identifiers 450K members, 34,000 at risk, $840K expiring in 30 days; do not infer results beyond it.
Before this case
Open Preview in a fresh conversation before running CLR-05. A previous response must not influence the evidence.
Pass CLR-06: Marketing Leader
Step 27 of 32
Action
Pass CLR-06: Marketing LeaderAs Marketing Leader, show me the top at-risk members and their profiles.
Expected resultThe captured Preview evidence records CLR-06 with the recorded identifiers Linda M., 12,400, $48K annual value; do not infer results beyond it.
Before this case
Open Preview in a fresh conversation before running CLR-06. A previous response must not influence the evidence.
Pass CLR-07: Marketing Leader
Step 28 of 32
Action
Pass CLR-07: Marketing LeaderAs Marketing Leader, create personalized win-back offers for each at-risk segment.
Expected resultThe captured Preview evidence records CLR-07 with the recorded identifiers Your favorite bags are 40% off, High-Value, 13,600; do not infer results beyond it.
Before this case
Open Preview in a fresh conversation before running CLR-07. A previous response must not influence the evidence.
Pass CLR-08: Marketing Leader
Step 29 of 32
Action
Pass CLR-08: Marketing LeaderAs Marketing Leader, get the win-back campaigns ready to launch and show me the expected results.
Expected resultThe captured Preview evidence records CLR-08 with the recorded identifiers Ready to launch, $489,600, 58:1; do not infer results beyond it.
Before this case
Open Preview in a fresh conversation before running CLR-08. A previous response must not influence the evidence.
Pass CLR-09: Marketing Leader
Step 30 of 32
Action
Pass CLR-09: Marketing LeaderAs Marketing Leader, how can we improve the overall loyalty program?
Expected resultThe captured Preview evidence records CLR-09 with the recorded identifiers Dynamic point expiry, +$340K/yr, 200 points from Gold; do not infer results beyond it.
Before this case
Open Preview in a fresh conversation before running CLR-09. A previous response must not influence the evidence.
Pass CLR-10: Marketing Leader
Step 31 of 32
Action
Pass CLR-10: Marketing LeaderAs Marketing Leader, summarize everything we accomplished in this loyalty session.
Expected resultThe captured Preview evidence records CLR-10 with the recorded identifiers Members analyzed, 34K ($2.1M points), $1.4M; do not infer results beyond it.
Before this case
Open Preview in a fresh conversation before running CLR-10. A previous response must not influence the evidence.
Confirm Draft and stop before publish
Step 32 of 32
Action
Confirm Draft and stop before publishExpected resultThe agent remains Draft and no Publish action is taken.
Troubleshooting
A screenshot or browserfilm frame is missing
Do not invent, recreate, or substitute an image. Keep the step open until its actual evidence is captured and reviewed. Remove --allow-pending only after all workshop evidence gates are met.
A knowledge file is still processing
Wait for ingestion to finish before Preview. A partial answer is not evidence.
A skill upload fails
Use the linked raw SKILL.md. Fix the reviewed source deliberately; do not silently skip the action.
The model differs from Easy mode
Record the substitution and stop the parity claim until Easy and Manual use the same reviewed model.
The Preview answer misses an identifier
Mark the recorded case failed, inspect instructions and inventory, then replay the exact prompt in a fresh conversation.
Should I publish?
No. Keep the same owned manual agent in Draft. Do not choose Publish as part of this tutorial.