Connect Withings Health Data to Claude

This MCP server integrates your Withings data into Claude, giving natural-conversation access to sleep, body measurements, workouts, heart rate data, and more.

Use it to query your own Withings health data conversationally - sleep quality over the past week, a weight trend over the last month, your average resting heart rate, whether you hit your weekly step goal, or comparing workout intensity between two months. It requires a Withings account with connected devices, and self-hosting requires a publicly accessible URL for OAuth callbacks - a hosted server is available at withings-mcp.com/mcp for anyone who doesn't want to deploy their own.

It exposes eleven tools: get_sleep_summary (sleep duration, phases - light/deep/REM, heart rate, breathing, sleep score), get_activity (daily steps, distance, calories, elevation, activity duration), get_intraday_activity (high-frequency activity data throughout the day), get_workouts (detailed workout data with heart rate zones and metrics), get_measures (weight, body composition, blood pressure, heart rate, temperature, VO2 max, and more), get_user_devices (connected Withings devices), get_user_goals (your health and fitness goals - steps, sleep, weight), list_heart_records and get_heart_signal (ECG record list and detailed ECG curve data), and list_stetho_records and get_stetho_signal (stethoscope record list and detailed sound signal data).

Beyond raw data access it analyzes sleep quality, duration, deep-sleep phases, and heart rate during sleep; tracks body metric trends including weight, body composition, and blood pressure; reviews workout and exercise patterns, calories burned, and heart rate zones; monitors heart health via ECG records; sets and tracks fitness and health goals; identifies patterns and correlations across sleep, activity, and other metrics; and generates AI-driven health trend analysis. It's built privacy-first - encrypted tokens, no logging of personal data, and secure OAuth 2.0 authentication with Withings.

To self-host from source:

git clone https://github.com/your-username/withings-mcp.git
cd withings-mcp
npm install
npm run generate-secret
cp .env.example .env
npm run build
npm start

Required environment variables: WITHINGS_CLIENT_ID and WITHINGS_CLIENT_SECRET (your Withings app credentials), WITHINGS_REDIRECT_URI (OAuth callback URL matching your Withings app settings), and ENCRYPTION_SECRET (32+ characters, generated via npm run generate-secret). Optional variables include PORT (default 3000), LOG_LEVEL (trace/debug/info/warn/error, default info), and ALLOWED_ORIGINS for browser-client CORS. All sensitive data is encrypted at rest with AES-256-GCM and stripped from logs automatically. Alternatively, use the hosted server by adding a custom connector in Claude Desktop's Settings pointed at the remote MCP URL, with no self-hosting required.

Withings device owners who want to query their own sleep, activity, heart health, and body composition data conversationally instead of digging through the Withings app.