Agent Example: Wellness Trend Analysis

Why This Matters

Wearable devices generate hundreds of data points per day, but raw numbers are overwhelming and clinically unhelpful. An agent that can:

• Detect meaningful trends in noisy time-series data
• Tie confidence to provenance — device-generated data with 150 samples/day gets higher confidence than sparse self-reported data
• Correlate across data streams — e.g., poor sleep correlating with elevated heart rate
• Write structured observations with proper provenance

...transforms raw device data into actionable clinical intelligence.

Wellness Data Available in the Reference Pod

The reference patient pod contains 30 days of data for Alex Rivera, 52M, Portland, OR:

Prerequisites

Option A: MCP via Claude Desktop

Step 1. Add the Cascade MCP server to your Claude Desktop configuration:

{
  "mcpServers": {
    "cascade": {
      "command": "cascade",
      "args": ["serve", "--mcp"],
      "env": {
        "CASCADE_POD_PATH": "/path/to/cascadeprotocol.org/reference-patient-pod"
      }
    }
  }
}

Step 2. Restart Claude Desktop, open a new conversation, and paste the contents of agent-prompt.md.

Step 3. The agent will use cascade_pod_query to read wellness data and cascade_write to write observations.

Option B: CLI-as-Skill via Claude Code

Step 1. Open a terminal in this directory and run Claude Code:

export CASCADE_POD_PATH="../reference-patient-pod"
claude

Step 2. Paste the contents of agent-prompt.md as your prompt. The agent will invoke the Cascade CLI:

# Read wellness time-series data
cascade pod query ./reference-patient-pod --wellness --json

# Or read individual categories
cascade pod query ./reference-patient-pod --heart-rate --json
cascade pod query ./reference-patient-pod --blood-pressure --json
cascade pod query ./reference-patient-pod --activity --json
cascade pod query ./reference-patient-pod --sleep --json

# Read clinical context (conditions, medications)
cascade pod query ./reference-patient-pod --conditions --json
cascade pod query ./reference-patient-pod --medications --json

# Write back observations
cascade pod write ./reference-patient-pod --type observation --json '{...}'

What the Agent Should Do

1. Read 30 days of wellness time-series data (heart rate, blood pressure, activity, sleep)
2. Identify clinically relevant trends (rising BP, declining activity, sleep irregularity)
3. Tie confidence to provenance — device data with many daily samples gets higher confidence than sparse self-reported data
4. Correlate across streams — link poor sleep to elevated heart rate, low activity to BP changes
5. Write trend observations back to the Pod with AIGenerated provenance

Files in This Example

Key Cascade Protocol Concepts Demonstrated