Guide sources and teaching inputs checked . Vendor directory claims retain their own review dates.
Use Maya’s teaching packet
Maya is an entirely fictional 54-year-old postmenopausal patient. Her goals are energy for work, time for her mother and staying strong. Use the downloadable S1–S8 packet for the first draft; add S9 only after you have reviewed it. No real patient information is needed for this exercise.
Ask for one HTML file that opens in a browser. Put the patient’s goals first, then a dated source timeline, relevant context, unresolved questions and a proposed follow-through table. Give every factual statement a source ID. Label inference separately from documented fact; leave clinical decisions for review.
Preserve the contradictions
The medication list includes atorvastatin, while a later message says Maya stopped it. Keep that discrepancy open for reconciliation. Four overnight CGM readings below 70 are unconfirmed sensor readings, not established hypoglycemia. Nine recorded nights and five missing nights are incomplete wearable data, not proof of five sleepless nights.
S9 adds caregiving four nights a week and charging the watch on some nights. This can change the feasibility of a plan and explain some missing data. It does not settle the medication question, the low readings or the cause of fatigue.
Review the file and request one change
Open the generated file, compare it with the packet and ask for one concrete revision. For each clinician-approved next step, identify an owner and review date. Keep the draft and approved version distinguishable.
This is a prepared case walkthrough. It does not connect to an EHR, perform clinical monitoring or require a live AI demonstration. Before adapting it to clinical use, evaluate the entire data path and test the intended integration.
Copy a prompt to try
Maya’s fictional living care map
Build & automate
Use only the fictional Maya S1–S8 teaching packet.
Create one HTML file I can open in a browser. Put her goals first.
Include a dated source timeline, context, uncertainties and a follow-through table.
Attach a source ID to every factual statement; label inference separately.
Keep the medication discrepancy, unconfirmed sensor lows and missing wearable nights open.
Do not diagnose, choose treatment or invent a clinician-approved plan.
Leave next-step decisions for clinician review, with owner and review-date fields.
After I review the file, I may provide S9. Show what S9 changes and what it does not resolve.
Use the entirely fictional packet linked from the workflows page. This exercise creates a teaching file; it does not connect to a clinical record.
This is an authored teaching example, not a captured AI output or a clinical plan.
Goal: Enough energy for work and helping her mother; stay strong. [S1]
Open question: The imported medication list includes atorvastatin [S4], while Maya reports stopping the cholesterol pill [S5]. Reconciliation remains open.
Data limit: Nine of fourteen wearable nights were recorded [S6]. Missing nights do not establish sleepless nights.
Review: Keep next-step decisions, owner and review date blank for clinician review.
Check before you use the result
Attach a source ID to each factual statement.
Keep the medication discrepancy and unconfirmed sensor lows open.
Treat missing wearable nights as missing data.
Add S9 after review and state which questions remain open.