Google Health MCP
Integrates with the Google Health API to read personal activity, sleep, workouts, body metrics, nutrition, devices, and account resources; generate daily and weekly/monthly health reports and TCX exports; and optionally create food, hydration, mood, symptom, and menstrual-period records with explicit confirmation.
This server does not currently have a deployable release on Glama. Follow the repository's setup instructions to install and configure it for your MCP client, including any required credentials and local dependencies.
Google Health MCP
A self-hosted, single-owner MCP server for Google Health, deployed on Cloudflare Workers. Read personal activity, sleep and health measurements; compare weekly trends; and optionally save food, water, mood, symptom and menstrual-period records to Google Health.
部署指南 / Deployment · Tool reference · Security · Changelog
Features
38 data types in a discoverable catalog, with supported filters, reconciliation, pagination and daily/physical-time aggregation. Skin-temperature sensors use the Google
v4betaendpoint.15 read tools: daily summaries, steps, sleep, workouts, body metrics, generic queries, aggregation, reports, nutrition summaries, workout details, TCX exports, devices, account resources and connection status.
Weekly/monthly reports compare equal-length periods, show coverage, and calculate sleep timing variation. Missing observations stay missing; failed queries and truncated results remain visible.
Optional
log_health_recordcreates food, hydration, mood, symptom or menstrual-period entries. It validates and previews the payload before an explicitly confirmed submission. Unknown calories can be omitted; the server does not invent nutrition facts.Optional Google-signed webhooks record change markers without retaining measurement payloads.
Owner-only Cloudflare Access login, OAuth consent, DCR and PKCE. One-hour access tokens; initial refresh grant lifetime of 90 days, renewed to a 90-day idle window on successful refresh.
这是可以实际写入 Google Health 的 MCP,饮食记录不只保存在聊天里。写入默认关闭,需要开启部署开关,并分别授予 Google 写入权限和 MCP health:write 权限。现有只读授权不会自动获得写权限。
Related MCP server: Google Health Broker
Quick start
Requirements: Node.js 22.18+, Python 3.10+, a Cloudflare account/domain, and a Google Cloud project enabled for Google Health API.
git clone https://github.com/p2o51/google-health-mcp.git
cd google-health-mcp
npm ci
cp .dev.vars.example .dev.vars
npm run authorize
npm run devLocal development uses an isolated loopback identity shortcut. Production requires a configured custom domain, a dedicated OAuth KV namespace, and Cloudflare Access protecting only /authorize. Complete the deployment guide before using a hosted client.
Examples of requests to your MCP client:
“Compare my last seven days with the previous seven days and show missing data.”
“Show my logged calories and water for yesterday.”
“Show the laps in my last run and export its TCX.”
“Log this lunch: tofu salad, 320 kcal and 22 g protein. Show me the entry first.”
Permissions and limits
Base Google scopes cover activity, sleep and health metrics. Nutrition, ECG, irregular-rhythm notifications, profile/settings and location exports require additional Google consent. Account/device availability still determines what data Google returns. See the scope matrix.
Food logging currently creates anonymous food entries with a display name and optional nutrition values. Google does not support editing those entries after creation. This MCP does not yet implement food-database search, update or delete tools. Mood/symptom/menstrual-period creation is supported; full-history reading for those types is not promised.
Reports are descriptive calculations from returned records, not Google/Fitbit proprietary readiness, sleep or stress scores. A webhook marks data changes; it does not send messages or run scheduled reports. This is a personal, single-account deployment, not a multi-tenant service.
Development
npm run typecheck
npm test
python3 scripts/authorize.py --all-read --writes nutrition,moods,symptoms,reproductive --print-scopesTests use synthetic health data and credentials. They cover protocol contracts, query limits, unit conversion, partial reports, write authorization, write previews, signed webhooks and OAuth refresh migration. No live health writes are required for the suite. Live API access and real-client OAuth consent must be verified separately on your own deployment.
The public configuration contains placeholders. .dev.vars, *.local.json, caches and build output are ignored. Never commit credentials, exported health records or your private deployment configuration.
References
Unofficial project; not affiliated with Google, Fitbit or Cloudflare. MIT licensed.
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
This server cannot be deployed
Maintenance
Related MCP Connectors
Your Google Health data (Fitbit, Pixel Watch): sleep, HRV, readiness and biological age.
Read-only Apple Health data synced from your iPhone: sleep, heart rate, HRV, workouts and trends.
Track, curate, and analyze data about your health, habits, and goals.
Household health records with source attribution. Read and add records with OAuth approval.
Related MCP Servers
- AlicenseAqualityBmaintenanceProvides read-only access to Withings health metrics including body composition, sleep, workouts, and ECG data with local SQLite caching and trend analysis. Features incremental synchronization, automatic OAuth token refresh, and supports all 200+ Withings measurement types for comprehensive health tracking.8100 PyPI1GPL 3.0
- AlicenseNot gradedqualityCmaintenanceSelf-hosted multi-user Google Health OAuth broker and read-only Agent API with remote MCP support. Enables users to connect their Google account and grant agents secure access to activity, sleep, heart, HRV, SpO2, and other health metrics via revocable per-token authentication.3MIT
- AlicenseNot gradedqualityBmaintenanceEnables AI assistants to read Fitbit/Google Health data (steps, sleep, heart rate, SpO2, workouts) and log weight, food, water, and exercises via natural language, self-hosted on Cloudflare Workers.5MIT
- AlicenseNot gradedqualityBmaintenanceEnables AI assistants such as Claude, ChatGPT and Gemini to read wearable health data — daily activity, sleep stages, heart rate, SpO2, HRV, workouts, weight and nutrition — through a self-hosted, OAuth-protected Cloudflare Worker endpoint. It also lets users log weight, body fat, water, food, exercises, mood and symptoms by voice or chat, while leaving sensor-recorded data read-only.MIT