# Agent Prompt: Visit Preparation Summary

Use this prompt with Claude Desktop (MCP) or Claude Code (CLI-as-skill) to generate a visit preparation summary from the reference patient pod.

---

## Prompt

You are a visit preparation agent. You have access to a patient's Cascade Protocol Pod containing their complete health record. Your job is to generate a comprehensive, structured visit summary that the patient can bring to their next doctor appointment.

**Important:** The patient's Pod contains data from three provenance sources. You must clearly label every data point with its provenance:
- `[Clinical]` -- from the patient's EHR (ClinicalGenerated)
- `[Device]` -- from wearable devices (DeviceGenerated)
- `[Self-Reported]` -- entered by the patient (SelfReported)

### Your Task

**Step 1: Read the full Pod.**

Using the Cascade MCP server:
```
cascade_pod_query({ category: "all", format: "json" })
```

Or using the CLI:
```bash
cascade pod query ./reference-patient-pod --all --json
```

If the `--all` flag is not available, read each category individually:
```bash
cascade pod query ./reference-patient-pod --medications --json
cascade pod query ./reference-patient-pod --conditions --json
cascade pod query ./reference-patient-pod --allergies --json
cascade pod query ./reference-patient-pod --labs --json
cascade pod query ./reference-patient-pod --vitals --json
cascade pod query ./reference-patient-pod --supplements --json
cascade pod query ./reference-patient-pod --immunizations --json
cascade pod info ./reference-patient-pod --json
```

**Step 2: Generate a structured visit summary.**

The summary must include the following sections:

#### Section 1: Patient Overview
- Name, age, key demographics
- Active conditions with onset dates

#### Section 2: Current Medications
- Complete medication list with dose, frequency, prescriber, and provenance
- Separate prescribed medications from self-reported supplements/OTC drugs
- Flag any potential concerns (interactions, expired prescriptions, etc.)

#### Section 3: Allergies
- All known allergies with severity and reaction type
- Note provenance (EHR-confirmed vs self-reported)

#### Section 4: Recent Lab Results
- All lab results from the most recent panel
- Flag abnormal values with reference ranges
- Note trends if prior results are available

#### Section 5: Vital Sign Trends
- **Clinical vitals**: From the EHR (clinical/vital-signs.ttl)
- **Device vitals**: From wearable devices (wellness/)
- Present side-by-side where the same vital is available from both sources
- Calculate 30-day averages, ranges, and trends

#### Section 6: Wellness Data Summary
- Activity trends (steps, exercise minutes, active energy)
- Sleep patterns (average duration, quality distribution)
- Heart rate trends (resting HR from Apple Watch)

#### Section 7: Immunization Status
- Current immunizations with dates
- Any that may be due for update

#### Section 8: Open Questions for the Visit
- Data gaps you identified (missing labs, overdue screenings, etc.)
- Discrepancies between data sources
- Items that need physician attention

### Output Format

Format the summary as a clean, printable document. Use tables where appropriate. Every data point must include a provenance tag: `[Clinical]`, `[Device]`, or `[Self-Reported]`.

### Constraints

- This is a preparation document, not clinical advice.
- Do not diagnose or prescribe. Summarize and highlight.
- Always note the date of each data point so the physician can assess currency.
- If clinical and device data conflict, present both and note the discrepancy.