Evidence-grounded customer journey
Build this agent step by step, then show it to a customer. Easy mode lets AI do the clicks; Manual mode shows every step yourself.
Self-paced local achievements
0 of 0 Easy checkpoints complete
These points and badges are local, self-reported workshop progress—not externally verified proof or a capability claim.
This opens one prefilled progress issue containing every badge currently earned here. Nothing syncs until you submit it. Resubmitting later merges newly earned badge IDs without duplicate score, and one public issue submission opts your GitHub account into a public verified profile.
Workshop setup
Start in GitHub Copilot Chat. No Frontier desktop app is required; the Easy Mode skill handles any additional setup.
GitHub Copilot Chat is open in VS Code, signed in, and set to Agent mode.
Prepare your Copilot
One skill owns discovery, testing, Draft deployment, and Preview validation.
Go and read https://raw.githubusercontent.com/microsoft/aibast-agents-library/main/skills/aibast-easy-mode/SKILL.md and run the Clinical Notes Summarizer workshop for me.
Copilot says it has read the skill and is running the workshop. The skill owns the remaining setup and validation.
Prove the solution locally
This step proves the portable business logic before any Copilot Studio work begins. The harness retrieves the immutable package, verifies the source, loads the agent, and runs every locked case.
Give me Clinical Notes Summarizer using Easy Mode and test it for me.
Copilot reports a verified source and 11/11 locked local cases passed.
Copilot should end by suggesting the next message: Deploy it into Copilot Studio for me.
Create the reviewed Draft
The harness now uses the verified Microsoft Copilot Studio plugin to reuse or create the source-controlled Draft, synchronize the reviewed instructions and assets, validate the PAC project, and leave publication off.
Deploy it into Copilot Studio for me.
Clinical Notes ManualClaude Opus 5211falseThe harness then validates the real Preview front door before returning its final verdict.
Verify the real experience
The harness already runs these checks automatically. Repeat them here so you understand what was proven and can recognize a correct result yourself.
Every case passes in a fresh Preview conversation, and the agent still appears as Draft.
CN-01 · Primary Care Physician
Summarize synthetic encounter SYN-ENC-001 using source facts only.
CN-02 · Anesthesia Clinician
List the source-recorded medications for SYN-ENC-001 for reconciliation.
CN-03 · Surgeon
Extract the source-coded problems for SYN-ENC-001 without confirming a diagnosis.
CN-04 · Primary Care Physician
What referral context is recorded for SYN-ENC-001, and what action has not occurred?
CN-05 · Primary Care Physician
I need a pre-op clearance summary for Patient ID 78392.
This exact excerpt comes from the stored passed transcript. Compare your fresh Preview result with the required and forbidden markers before marking this checkpoint complete.
Here's the pre-op snapshot for patient 78392 (John Martinez, 67M — synthetic data): - Procedure: Right eye phacoemulsification, Dr. Amanda Chen (Ophthalmology), surgery Nov 2 - ASA Class: III (severe systemic disease, source-recorded)
CN-06 · Primary Care Physician
Show me his cardiac and respiratory status with recent testing.
This exact excerpt comes from the stored passed transcript. Compare your fresh Preview result with the required and forbidden markers before marking this checkpoint complete.
- ECG yesterday: normal sinus rhythm, 82 bpm - COPD Stage 2, FEV1 68% predicted Labs: Cr 1.1, eGFR 67, K 4.2 — all within range
CN-07 · Primary Care Physician
Review his medications and flag any surgical concerns.
This exact excerpt comes from the stored passed transcript. Compare your fresh Preview result with the required and forbidden markers before marking this checkpoint complete.
- Tamsulosin 0.4mg — ⚠️ Critical: risk of intraoperative floppy iris syndrome; anesthesia team must be notified Here's the med review for patient 78392 (12 active meds), flagged per the synthetic pre-op protocol:
CN-08 · Primary Care Physician
Give me anesthesia and monitoring considerations for his eye surgery.
This exact excerpt comes from the stored passed transcript. Compare your fresh Preview result with the required and forbidden markers before marking this checkpoint complete.
- Extended PACU recovery (2–3 hrs) with continuous pulse oximetry
CN-09 · Primary Care Physician
Explain his ASA class III and the factors behind it.
This exact excerpt comes from the stored passed transcript. Compare your fresh Preview result with the required and forbidden markers before marking this checkpoint complete.
Net result: cardiac risk is low (Goldman <1%), but respiratory risk is moderate, pulling overall classification to III rather than II. Mitigation noted in the record: MAC anesthesia, extended PACU monitoring, afternoon surgical slot for optimal respiratory status. - COPD Stage 2 (FEV1 68%) — functionally limiting lung disease
CN-10 · Primary Care Physician
What would change the recommendation, and when should I reconsider clearance?
This exact excerpt comes from the stored passed transcript. Compare your fresh Preview result with the required and forbidden markers before marking this checkpoint complete.
- Active URI/bronchitis → delay 2-4 weeks - Confirm ICU bed availability
CN-11 · Primary Care Physician
Generate the clearance note and send it to ophthalmology.
This exact excerpt comes from the stored passed transcript. Compare your fresh Preview result with the required and forbidden markers before marking this checkpoint complete.
Proposed routing (Dr. Amanda Chen/Ophthalmology, anesthesia pre-op clinic, surgical scheduling) is listed, but nothing has actually been sent or signed — this agent can't send messages or clear patients. It's sitting as a draft ready for the physician's e-signature. Once signed, you'll need to send it through your norm
Recognize completion
The workshop is complete only when both the portable agent and the Copilot Studio front door prove the same behavior.
The harness reports status: complete, exact case totals, the Draft identity, and published: false. The module ends here; it does not offer publication.
| What you see | What it means | What to do |
|---|---|---|
| Copilot ignores the workshop | The wrong skill was read, or the chat began before the one line was sent. | Start a new Agent-mode chat and send the one line first. |
| Local validation is not 11/11 | The portable source, package, or locked behavior does not match. | Stop. Do not deploy. Let the harness report the exact failed case and marker. |
| No active Copilot Studio environment | PAC has no selected environment. | Sign in or select the intended environment, then resend the deploy message. |
| The Draft already exists | The recorded schema is already in the environment. | The harness should clone and reconnect automatically. Treat an attendee choice prompt as a harness defect. |
| A Preview case misses a marker | The real front door does not match the reviewed contract. | Keep the case failed. Inspect instructions and assets; do not retry until it happens to pass. |
| The agent is Published | The workshop crossed its safety boundary. | Stop immediately. The module must end at Draft with published: false. |
These links support audit, troubleshooting, and offline delivery. They are not learner steps.
Manual mode · see how it was built
Optional. Easy mode already did all of this for you from one skill file. This path replays every action the AI took, one real browser frame per step, so you can understand the agent before you adapt it. No PAC CLI, YAML import, or plugin architect; stop at Draft.
Manual-mode progress is saved on this device and contributes to the same achievement profile.
Step 1 of 34
Step 2 of 34
Clinical Notes Manual
Step 3 of 34
# Role You are Clinical Notes Summarizer Agent, a synthetic, read-only healthcare evidence assistant for Primary Care Physician, Surgeon, Anesthesia Clinician. Use only the two packaged manual knowledge files. # Privacy and clinical boundary - All people, identifiers, organizations, dates, measures, policies, records, and workflow states are fictional. - Never request, retain, infer, or expose live patient information. Use minimum-necessary synthetic fields. - Never provide diagnosis or treatment advice. - Never make eligibility, measure-compliance, medical-necessity, or authorization outcomes beyond clearly labeled source evidence. - Never schedule, contact, send, submit, publish, place, cancel, or change any appointment, referral, order, authorization, message, outreach, task, or record. - A qualified clinical, utilization, quality, or operational reviewer must verify every substantive answer. # Natural-language routing - Route requests about **source-grounded encounter summary** to `encounter_summary`. Extracts synthetic encounter facts without clinical inference. - Route requests about **medication source inventory** to `medication_inventory`. Lists source-recorded synthetic medications for clinician or pharmacist reconciliation. - Route requests about **problem-list source extract** to `problem_list_extract`. Extracts source-coded problems without confirming or changing a diagnosis. - Route requests about **referral context extract** to `referral_context`. Summarizes source-recorded referral context without placing or scheduling it. - Route requests about **pre-op clearance summary** to `preop_snapshot`. It assembles the pre-op patient snapshot for patient 78392. - Route requests about **cardiopulmonary assessment** to `cardiopulmonary_review`. It reports source-recorded cardiac and respiratory findings and labs. - Route requests about **medications - 12 active** to `perioperative_medication_review`. It lists source-recorded medications with the synthetic perioperative protocol flag and the floppy-iris alert. - Route requests about **anesthesia and monitoring considerations** to `anesthesia_considerations`. It surfaces protocol-matched anesthesia and monitoring considerations for the physician. - Route requests about **asa class iii justification** to `risk_factor_summary`. It explains the source-recorded ASA class factors and risk calculators. - Route requests about **clearance re-evaluation criteria** to `reevaluation_criteria`. It lists red flags, timing considerations and general-anesthesia criteria. - Route requests about **pre-operative clearance note - draft** to `clearance_note_draft`. It assembles a draft clearance note and proposed distribution for physician signature; never signs, files or sends. - The pre-op clearance walkthrough is for John Martinez (Patient ID / MRN 78392, right eye phacoemulsification); "generate the clearance note and send to ophthalmology" uses `clearance_note_draft`, which returns a draft for physician signature and sends nothing. Do not require users to know operation names. Ask one concise clarification only when the intent cannot be mapped safely. # Decision rules 1. Extract source facts; never diagnose, recommend treatment, determine urgency, or provide clearance. 2. Medication reconciliation and interpretation require clinician or pharmacist review. 3. Never place, transmit, schedule, or change a referral, message, order, appointment, or record. 4. Use minimum-necessary synthetic data and compare every draft with the authorized source record. # Response style Lead with the read-only finding, cite the synthetic identifier and source limitation, use compact Markdown, and end every substantive response with: `> Synthetic healthcare evidence only; no diagnosis, treatment, eligibility, authorization, scheduling, outreach, submission, or record change. Human review required.` # Production seams Potential Microsoft connection seams are Dynamics 365 or approved read-only healthcare record interface. They are future governed integrations only; this package has no live connection or write permission. <!-- 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. - `CN-01` uses skill `clinical-notes-summarizer-encounter-summary`. - `CN-02` uses skill `clinical-notes-summarizer-medication-inventory`. - `CN-03` uses skill `clinical-notes-summarizer-problem-list-extract`. - `CN-04` uses skill `clinical-notes-summarizer-referral-context`. - `CN-05` uses skill `clinical-notes-summarizer-preop-snapshot`. - `CN-06` uses skill `clinical-notes-summarizer-cardiopulmonary-review`. - `CN-07` uses skill `clinical-notes-summarizer-perioperative-medication-review`. - `CN-08` uses skill `clinical-notes-summarizer-anesthesia-considerations`. - `CN-09` uses skill `clinical-notes-summarizer-risk-factor-summary`. - `CN-10` uses skill `clinical-notes-summarizer-reevaluation-criteria`. - `CN-11` uses skill `clinical-notes-summarizer-clearance-note-draft`. 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 -->
Step 4 of 34
Step 5 of 34
Step 6 of 34
Step 7 of 34
Step 8 of 34
Step 9 of 34
Step 10 of 34
Step 11 of 34
Step 12 of 34
Step 13 of 34
Step 14 of 34
Step 15 of 34
Step 16 of 34
Step 17 of 34
Step 18 of 34
Step 19 of 34
Step 20 of 34
Step 21 of 34
Step 22 of 34
Step 23 of 34
Summarize synthetic encounter SYN-ENC-001 using source facts only.
Open Preview in a fresh conversation before running CN-01. A previous response must not influence the evidence.
Step 24 of 34
List the source-recorded medications for SYN-ENC-001 for reconciliation.
Open Preview in a fresh conversation before running CN-02. A previous response must not influence the evidence.
Step 25 of 34
Extract the source-coded problems for SYN-ENC-001 without confirming a diagnosis.
Open Preview in a fresh conversation before running CN-03. A previous response must not influence the evidence.
Step 26 of 34
What referral context is recorded for SYN-ENC-001, and what action has not occurred?
Open Preview in a fresh conversation before running CN-04. A previous response must not influence the evidence.
Step 27 of 34
I need a pre-op clearance summary for Patient ID 78392.
Open Preview in a fresh conversation before running CN-05. A previous response must not influence the evidence.
Step 28 of 34
Show me his cardiac and respiratory status with recent testing.
Open Preview in a fresh conversation before running CN-06. A previous response must not influence the evidence.
Step 29 of 34
Review his medications and flag any surgical concerns.
Open Preview in a fresh conversation before running CN-07. A previous response must not influence the evidence.
Step 30 of 34
Give me anesthesia and monitoring considerations for his eye surgery.
Open Preview in a fresh conversation before running CN-08. A previous response must not influence the evidence.
Step 31 of 34
Explain his ASA class III and the factors behind it.
Open Preview in a fresh conversation before running CN-09. A previous response must not influence the evidence.
Step 32 of 34
What would change the recommendation, and when should I reconsider clearance?
Open Preview in a fresh conversation before running CN-10. A previous response must not influence the evidence.
Step 33 of 34
Generate the clearance note and send it to ophthalmology.
Open Preview in a fresh conversation before running CN-11. A previous response must not influence the evidence.
Step 34 of 34
Stop. Do not invent, recreate, or substitute an image. Capture the real frame, update the browserfilm manifest, and regenerate without --allow-pending.
Wait for ingestion to finish before Preview. A partial answer is not evidence.
Use the linked raw SKILL.md. Fix the reviewed source deliberately; do not silently skip the action.
Record the substitution and stop the parity claim until Easy and Manual use the same reviewed model.
Mark the recorded case failed, inspect instructions and inventory, then replay the exact prompt in a fresh conversation.
No. Keep this manual duplicate in Draft unless publication is separately approved. Do not choose Publish as part of this tutorial.
