Sandbox
@akutishevsky/withings-mcp

MCP server for Withings health data

This repo is a remote MCP connector that brings Withings health data into an MCP client. It handles OAuth, stores tokens in Supabase, and exposes tools for sleep, activity, measures, workouts, and heart records.

42 stars9 forksTypeScriptUpdated 8d ago
Who it's for

Builders who want their agent to answer questions from their Withings account.

What it delivers

You can ask an MCP client about your sleep, workouts, and body metrics instead of checking Withings manually.

What it does

OAuth connection to Withings

Authenticates a Withings account through OAuth 2.0 with redirect handling and PKCE support.

Health data tools

Provides MCP tools for sleep summaries, activity, intraday activity, workouts, measures, devices, goals, ECG records, and stethoscope data.

Encrypted token storage

Encrypts access tokens, refresh tokens, and auth codes before storing them in the database.

Privacy-safe logging

Redacts tokens, user IDs, and other sensitive values from logs.

Supabase-backed session storage

Uses Supabase migrations and database tables for sessions, tokens, and cleanup.

How to get it

  1. 1Make sure your redirect URI in the .env file matches the publicly accessible URL…
    # The project runs TypeScript directly with Bun — no build step required.
    bun run start
  2. 2Discovery URL: /.well-known/oauth-authorization-server
    bun run start            # Run the server
    bun run dev              # Hot-reload mode
    bun run typecheck        # Type-check with tsc (no emit)
    bun run build            # Bundle for production (outputs to ./build)
    bun run generate-secret  # Generate encryption secret for ENCRYPTION_SECRET env variable

README

Withings MCP Server

Help me pay for the servers on Patreon

Quality Gate Status Bugs Code Smells Lines of Code Reliability Rating Security Rating Maintainability Rating Vulnerabilities

A Model Context Protocol (MCP) server that brings your Withings health data into Claude. Access your sleep patterns, body measurements, workouts, heart data, and more through natural conversation.

🔒 Privacy First: This is my personal project, and the repository is intentionally public to demonstrate transparency. The code shows that no personal information is logged or stored maliciously. All sensitive data (tokens, user IDs) is encrypted at rest and automatically redacted from logs. You can review the entire codebase to verify this commitment to privacy.

⚠️ Disclaimer: This server is provided as-is without any guarantees or warranties. While I've made every effort to ensure security and privacy, I make no guarantees about availability, data integrity, or security. Use at your own risk. For production use cases, consider self-hosting your own instance.

Quick Setup

  1. Open Claude DesktopCustomizeConnectors
  2. Click +Add Custom Connector
  3. Set URL to https://withings-mcp.com/mcp → click Add
  4. Click Connect and authorize with your Withings account

That's it! Ask Claude about your sleep, weight, workouts, or heart data.

Demo

Demo

Table of Contents

What Can You Do With This?

This MCP server gives Claude access to your Withings health data, allowing you to:

  • Analyze your sleep: Ask about sleep quality, duration, deep sleep stages, heart rate during sleep
  • Track body metrics: Weight trends, body composition, blood pressure, heart rate over time
  • Review workouts: Analyze exercise patterns, calories burned, heart rate zones
  • Monitor heart health: Access ECG recordings and detailed heart data
  • Set and track goals: Review your fitness and health goals
  • Identify patterns: Find correlations between sleep, activity, and other metrics
  • Generate insights: Get AI-powered analysis of your health trends

All through natural conversation with Claude or any other MCP-compatible client.

For End Users: Using the Hosted Server

If you just want to use this MCP server with Claude Desktop without hosting anything yourself, follow these steps:

Prerequisites

  1. A Withings account with connected devices
  2. Claude Desktop or any other MCP-compatible client installed on your computer

Setup Instructions

Step 1: Add Connector in Claude Desktop

  1. Open Claude Desktop
  2. Go to Customize (in the sidebar or menu)
  3. Navigate to the Connectors section
  4. Click the + button to add a new connector
  5. Select Add Custom Connector
  6. Fill in the following details:
    • Name: Withings (or any name you prefer)
    • Remote MCP server URL: https://withings-mcp.com/mcp
  7. Click Add

Note: If your MCP client doesn't support UI-based connector configuration, you can manually edit the config file instead. See the manual configuration guide below.

Step 2: Connect and Authorize

  1. In the Connectors settings, find the Withings connector you just added
  2. Click Connect next to the connector
  3. Your web browser will open with the Withings authorization page
  4. Log in to your Withings account
  5. Review and approve the permissions requested
  6. You'll be redirected back and the connection will be complete

After authorization, Claude will have access to your Withings data!

Available Tools

Once connected, Claude can use these tools to access your data:

Sleep & Activity

  • get_hrv - Minute-level RMSSD, SDNN, and quality scores captured during sleep
  • get_sleep_summary - Sleep duration, stages (light/deep/REM), heart rate, breathing, sleep score
  • get_activity - Daily steps, distance, calories, elevation, activity durations
  • get_intraday_activity - High-frequency activity data throughout the day
  • get_workouts - Detailed workout summaries with heart rate zones and metrics

Body Measurements

  • get_measures - Weight, body composition, blood pressure, heart rate, temperature, VO2 max, and more

Devices & Goals

  • get_user_devices - List of connected Withings devices
  • get_user_goals - Your health and fitness goals (steps, sleep, weight)

Heart Health

  • list_heart_records - List of ECG recordings
  • get_heart_signal - Detailed ECG waveform data

Stethoscope (if you have BPM Core)

  • list_stetho_records - List of stethoscope recordings
  • get_stetho_signal - Detailed audio signal data

Example Conversations

Try asking Claude:

  • "How has my sleep quality been over the past week?"
  • "Show me my weight trend for the last month"
  • "What's my average resting heart rate?"
  • "Did I hit my step goal this week?"
  • "Compare my workout intensity between this month and last month"
  • "When did I sleep best this month?"

Privacy & Security

  • Encrypted tokens: All authentication tokens and authorization codes are encrypted using AES-256-GCM before storage
  • No logging of personal data: The code is public - you can verify that no sensitive information is logged
  • Automatic redaction: All user IDs, tokens, and credentials are automatically redacted from system logs
  • OAuth 2.0: Industry-standard secure authentication with PKCE support and redirect URI validation
  • Session security: MCP sessions are bound to the authenticated user, preventing cross-user access
  • You're in control: Revoke access anytime from your Withings account settings

For Developers: Self-Hosting

Want to run your own instance? Here's how to deploy this MCP server yourself.

Prerequisites

  1. Bun 1.1+ installed
  2. A Withings Developer Account

Step 1: Create Withings Application

  1. Go to Withings Developer Portal
  2. Create a new application
  3. Note your Client ID and Client Secret
  4. Set your Redirect URI to: https://your-domain.com/callback
    • This must be a publicly accessible URL (localhost is not supported by Withings)
    • Can be any domain where you'll host the server (e.g., Fly.io, Railway, your own server, etc.)

Important: Remove Google Analytics

The hosted version includes a Google Analytics tag (G-ZMGF9WXL3W) in the static pages under public/. If you're forking this repo, remove or replace the GA snippet in public/index.html and public/health.html, and update the CSP headers in src/server/app.ts accordingly.

Step 2: Clone and Setup

# Clone the repository
git clone https://github.com/your-username/withings-mcp.git
cd withings-mcp

# Install dependencies
bun install

# Generate encryption secret
bun run generate-secret
# Copy the output - you'll need it for environment variables

Step 2.5: Set Up Supabase Database

  1. Create a free project at Supabase
  2. Install the Supabase CLI: bun install -g supabase (or use brew install supabase/tap/supabase)
  3. Link your project: supabase link --project-ref <your-project-ref>
  4. Apply the database migrations: supabase db push
  5. Get your credentials from Dashboard → Settings → API:
    • Project URLSUPABASE_URL
    • Service role keySUPABASE_SECRET_KEY

Step 3: Local Development

Note: Withings requires a publicly accessible URL for OAuth callbacks. For local development, use a tunneling service to expose your local server or deploy to a staging environment for testing.

# Copy environment template
cp .env.example .env

# Edit .env with your values
# WITHINGS_CLIENT_ID=your_client_id
# WITHINGS_CLIENT_SECRET=your_client_secret
# WITHINGS_REDIRECT_URI=https://your-tunnel-url.com/callback
# ENCRYPTION_SECRET=paste_generated_secret_here
# SUPABASE_URL=https://your-project.supabase.co
# SUPABASE_SECRET_KEY=your_service_role_key
# PORT=3000

# Run locally (Bun executes TypeScript directly — no build step)
bun run dev

Make sure your redirect URI in the .env file matches the publicly accessible URL pointing to your local server.

Step 4: Deploy to Production

# The project runs TypeScript directly with Bun — no build step required.
bun run start

Deploy to DigitalOcean App Platform (its Bun buildpack detects package.json and runs bun run start automatically), or any other host that supports Bun.

Set the following environment variables on your hosting platform:

VariableRequiredExample
WITHINGS_CLIENT_IDYesyour_client_id
WITHINGS_CLIENT_SECRETYesyour_client_secret
WITHINGS_REDIRECT_URIYeshttps://your-domain.com/callback
ENCRYPTION_SECRETYesGenerated from step 2
SUPABASE_URLYeshttps://your-project.supabase.co
SUPABASE_SECRET_KEYYesYour Supabase service role key
PORTNo3000 (or your platform's default)
LOG_LEVELNoinfo
ALLOWED_ORIGINSNohttps://example.com,https://app.example.com

Step 5: Update Withings App Settings

Go back to your Withings developer app and update the redirect URI to match your deployed URL: https://your-domain.com/callback

Step 6: Configure Your MCP Client

For Claude Desktop:

  1. Open Claude Desktop
  2. Go to CustomizeConnectors section
  3. Click the + button, then select Add Custom Connector
  4. Fill in the following details:
    • Name: Withings (or any name you prefer)
    • Remote MCP server URL: https://your-domain.com/mcp
  5. Click Add
  6. Click Connect next to the connector to authorize

For Other MCP Clients:

Configure your MCP client with the following connection details:

  • Server URL: https://your-domain.com
  • Transport: Streamable HTTP
  • Endpoint: /mcp
  • Authentication: OAuth 2.0
  • Discovery URL: /.well-known/oauth-authorization-server

Environment Variables Reference

VariableRequiredDescription
WITHINGS_CLIENT_IDYesYour Withings app client ID
WITHINGS_CLIENT_SECRETYesYour Withings app client secret
WITHINGS_REDIRECT_URIYesOAuth callback URL (must match Withings app settings)
ENCRYPTION_SECRETYes32+ character secret for token encryption (generate with bun run generate-secret)
SUPABASE_URLYesYour Supabase project URL (from Dashboard → Settings → API)
SUPABASE_SECRET_KEYYesYour Supabase service role key (from Dashboard → Settings → API)
PORTNoServer port (default: 3000)
LOG_LEVELNoLogging level: trace, debug, info, warn, error (default: info)
ALLOWED_ORIGINSNoComma-separated list of allowed CORS origins for browser clients

Development Commands

bun run start            # Run the server
bun run dev              # Hot-reload mode
bun run typecheck        # Type-check with tsc (no emit)
bun run build            # Bundle for production (outputs to ./build)
bun run generate-secret  # Generate encryption secret for ENCRYPTION_SECRET env variable

Project Structure

src/
├── auth/              # OAuth 2.0 authentication & token storage
├── db/                # Supabase client & cleanup scheduler
├── server/            # Hono app, MCP endpoints, middleware
├── tools/             # MCP tools for Withings API (sleep, measure, user, heart, stetho)
├── types/             # TypeScript type definitions (Hono, Withings API)
├── withings/          # Withings API client
├── utils/             # Logger and encryption utilities
└── index.ts           # Main entry point

supabase/
└── migrations/        # Database schema migrations

See CLAUDE.md for detailed architecture documentation.

Security Features

Token Encryption

All Withings access tokens, refresh tokens, and authorization codes are encrypted at rest using AES-256-GCM:

  • Algorithm: AES-256-GCM (authenticated encryption)
  • Key Derivation: PBKDF2 with 100,000 iterations
  • Defense in Depth: Even if the database is compromised, tokens remain protected

Important: Keep your ENCRYPTION_SECRET:

  • At least 32 characters long
  • Randomly generated (use bun run generate-secret)
  • Secure and never committed to version control
  • Consistent across server restarts

OAuth Hardening

  • Redirect URI validation: The /authorize endpoint validates redirect_uri against the registered client's allowed URIs, preventing open redirect attacks
  • Single-use auth codes: Authorization codes are atomically consumed to prevent replay attacks (per RFC 6749)
  • PKCE support: SHA-256 code challenge method for enhanced security
  • Startup validation: Server refuses to start if required environment variables are missing

Transport Security

  • Session-token binding: MCP sessions are bound to the bearer token that created them, preventing cross-user session hijacking
  • JSON-RPC validation: All incoming messages are validated against the JSON-RPC 2.0 specification before processing
  • Request body limits: 1MB global limit to prevent memory exhaustion
  • HTTPS redirect: HTTP requests are automatically redirected to HTTPS in production
  • Strict CSP: Content Security Policy with no unsafe-inline directives
  • Atomic rate limiting: PostgreSQL function with row-level locking prevents race conditions

Privacy-Safe Logging

The custom logger automatically redacts all sensitive information:

  • ✅ Operational events and errors logged
  • ❌ No tokens, credentials, or auth codes
  • ❌ No user IDs or personal information
  • ❌ No API request/response payloads with sensitive data

You can review the logging implementation in src/utils/logger.ts.

Contributing

This is a personal project, but contributions are welcome! Please:

  1. Fork the repository
  2. Create a feature branch
  3. Make your changes
  4. Submit a pull request

License

MIT License - see LICENSE file for details.

Support

Acknowledgments

Built with:

Files in the repo

Repository payload15 top-level entries
  • .github
  • demo
  • public
  • src
  • supabase
  • tests
  • .env.example
  • .gitignore
  • bun.lock
  • CLAUDE.md
  • LICENSE
  • package.json
  • README.md
  • server.json
  • tsconfig.json

Discussion (0)

Ask about usage, or say what you built with it

Sign in to join the discussion.

No comments yet. Be the first to say what this is good for.

More connectors

Real-time global intelligence dashboard. AI-powered news aggregation, geopolitical monitoring, and infrastructure tracking in a unified situational awareness interface

86k

High-performance code intelligence MCP server. Indexes codebases into a persistent knowledge graph — average repo in milliseconds. 158 languages, sub-ms queries, 99% fewer tokens. Single static binary, zero dependencies.

43k

Universal provider proxy for OpenAI Codex & Claude Code — use any LLM (Claude, Gemini, Grok, DeepSeek, Ollama…) with Codex CLI, App, SDK, and Claude Code

14k
okf-memory/
okf-agent-memory

Git-native persistent memory for AI coding agents. Implements Google OKF v0.2 with sub-300µs in-memory BM25 search, embedded MCP server, and progressive disclosure. Slashes token bloat by 80% with zero external databases or dependencies. Built in pure Go.

547
2akouwu/
reverify

Stop your AI from making things up — it proposes, deterministic tools decide, every claim checked against ground truth with evidence. Grounded facts and context survive resets. Reverse engineering is the proving ground. MCP server + CLI.

1.1k
t8y2/dbxConnectors

20 MB lightweight cross-platform database client for 90+ databases, including MySQL, PostgreSQL, SQLite, Redis, MongoDB, DuckDB, SQL Server, and Dameng. Built-in AI, MCP Server, CLI, desktop and Docker. | 轻量级跨平台数据库管理工具,支持 MySQL、PostgreSQL、SQLite、Redis、MongoDB、达梦等 90+ 数据库,提供桌面端、Docker、CLI、内置 AI 助手和 MCP Server。

19k