Whoop MCP Server
Enables users to access Whoop biometrics including recovery, sleep, and strain scores while facilitating activity logging and device management. It allows for seamless interaction with health data to set alarms, update weight, and track activities through natural language commands.
README
whoop-write-mcp
MCP server for Whoop — read biometrics, log activities, and manage your Whoop band from Claude.
Built on whoop-write-api and the Model Context Protocol.
Tools
Read
| Tool | Description |
|---|---|
get_recovery |
Recovery score, HRV, resting HR (supports date range) |
get_sleep |
Sleep score, hours, respiratory rate |
get_strain |
Daily strain and heart rate from cycles |
get_body_measurement |
Height, weight, max HR |
get_sport_types |
List all Whoop sport/activity type IDs |
search_exercises |
Search exercise catalog by name, equipment, muscle group |
get_journal_behaviors |
Available journal questions for a date |
Write
| Tool | Description |
|---|---|
create_activity |
Log sauna, meditation, yoga, running, etc. |
delete_activity |
Remove an activity by ID |
list_activities |
List recent activities with IDs |
update_weight |
Update body weight (lbs, converted to kg) |
set_alarm |
Set or disable Whoop alarm |
log_workout |
Full workout with optional exercise detail |
link_exercises |
Attach per-set exercise data to an activity |
log_journal |
Log caffeine, alcohol, supplements, notes |
Connect to Claude Code
Step 1: Add the server
claude mcp add -s user whoop -- uvx whoop-write-mcp
Step 2: Authenticate
whoop-write-mcp login
Enter your Whoop email and password when prompted. Credentials are used once to get tokens, then discarded. Tokens are cached at ~/.whoop/tokens.json and auto-refresh — you shouldn't need to login again.
Step 3: Use it
Ask Claude things like:
- "What's my recovery today?"
- "Log a 20-minute sauna session"
- "Set my alarm for 7:30 AM"
- "How did I sleep last night?"
- "Update my weight to 255 lbs"
- "Delete my last activity"
Alternative: env vars (skip the login step)
claude mcp add -e WHOOP_EMAIL=you@example.com -e WHOOP_PASSWORD=yourpass -s user whoop -- uvx whoop-write-mcp
Server auto-authenticates on first tool call. Useful for automation.
From source
git clone https://github.com/jd1207/whoop-write-mcp.git
cd whoop-write-mcp && pip install -e .
claude mcp add -s user whoop -- whoop-write-mcp
whoop-write-mcp login
Managing
claude mcp list # see registered servers
claude mcp remove whoop # unregister
whoop-write-mcp status # check auth state
whoop-write-mcp logout # remove cached tokens
Or type /mcp inside Claude Code to manage connected servers.
How Auth Works
- First tool call checks for cached tokens at
~/.whoop/tokens.json - No tokens? Auto-authenticates from
WHOOP_EMAIL/WHOOP_PASSWORDenv vars if set - No env vars? Returns a clear error: "Run
whoop-write-mcp loginin your terminal" - Tokens cached (
0600permissions), auto-refresh on expiry - Password never stored on disk — only refresh tokens
Architecture
Claude Code <--stdio--> whoop-write-mcp <--https--> Whoop API
|
~/.whoop/tokens.json
Claude Code Skill
The skill/ directory contains a coaching intelligence skill that teaches Claude how to interpret Whoop data and make training decisions — not just call tools, but use them like a coach.
Install it to make any Claude Code agent smarter with Whoop data:
cp -r skill/ ~/.claude/skills/whoop-mcp/
What the skill provides:
- Decision frameworks — when to train heavy, when to deload, when to stay quiet
- Signal priority — sleep > HRV trend > recovery score > strain
- Verbosity rules — only surface data when it changes the recommendation
- Multi-step workflows — morning check-in, pre-workout assessment, post-workout sync
- Biometric interpretation — recovery zones, HRV trends, sleep thresholds, strain budgets
Related Projects
- whoop-write-api — the underlying Python library for Whoop's reverse-engineered API
- SpotMe — self-hosted AI workout coach PWA that uses this MCP server
Development
git clone https://github.com/jd1207/whoop-write-mcp.git
cd whoop-write-mcp
pip install -e ".[dev]"
pytest
License
MIT
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.