whoop-mcp-server
MCP server that exposes Whoop API v2 data as tools, enabling natural language queries for cycles, recovery, sleep, workouts, and profile.
README
whoop-mcp-server
MCP server for the official Whoop API v2, written in Go. Exposes cycles, recovery, sleep, workouts, and profile data as MCP tools.
Tools
All tools return JSON. Record shapes are the unmodified Whoop API v2 objects, so every field the API exposes is available to the model.
| Tool | Inputs | Returns |
|---|---|---|
get_profile |
— | Single object: user_id, email, first_name, last_name |
get_body_measurement |
— | Single object: height_meter, weight_kilogram, max_heart_rate |
get_cycles |
start_date, end_date, limit (all optional) |
Array of cycle records — one per physiological day: start/end, score_state, and score with day strain, kilojoule, average_heart_rate, max_heart_rate |
get_latest_cycle |
— | Single cycle record (the current/most recent day) |
get_recoveries |
start_date, end_date, limit (all optional) |
Array of recovery records — one per sleep: cycle_id, sleep_id, and score with recovery_score (0–100), resting_heart_rate, hrv_rmssd_milli, spo2_percentage, skin_temp_celsius |
get_recovery_for_cycle |
cycle_id (required) |
Single recovery record for that cycle |
get_sleep |
start_date, end_date, limit (all optional) |
Array of sleep records: start/end, nap flag, and score with per-stage durations (light/SWS/REM/awake, in ms), respiratory_rate, sleep_performance_percentage, sleep_efficiency_percentage, sleep-need breakdown |
get_workouts |
start_date, end_date, limit (all optional) |
Array of workout records: sport_name, start/end, and score with workout strain, heart rates, kilojoule, distance_meter, altitude_gain_meter, time-in-zone durations (zone_durations, ms per HR zone) |
get_average_strain |
days (optional, default 7) |
Computed aggregate: {days, cycles_counted, average_strain} |
check_auth_status |
— | {authenticated: bool} plus the profile on success or an error string on failure |
Notes:
- Dates are
YYYY-MM-DD;end_dateis inclusive (internally mapped to the start of the next day, per the API's exclusiveendsemantics). - Collections are date-descending (most recent first) and paginated transparently — the server follows
next_tokenuntillimitrecords (default 25, max 100) are collected. - Scores are point-in-time per record, not time-series: one strain value per cycle, one recovery per sleep. For trends, fetch a range and let the model aggregate (or use
get_average_strainfor the built-in strain average). - A record's
score_statecan beSCORED,PENDING_SCORE, orUNSCORABLE—scoreis only present whenSCORED.
Setup
1. Create a Whoop developer app
At developer.whoop.com, create an app with:
- Redirect URI:
http://localhost:8719/callback - Scopes:
read:cycles,read:recovery,read:sleep,read:workout,read:profile,read:body_measurement,offline
Note the client ID and client secret.
2. Install
Homebrew:
brew install zayden16/tap/whoop-mcp-server
Or download a prebuilt binary from Releases, or:
go install github.com/Zayden16/whoop-mcp-server@latest
Or build from source:
go build -o whoop-mcp-server .
3. Authorize (one time)
export WHOOP_CLIENT_ID=...
export WHOOP_CLIENT_SECRET=...
./whoop-mcp-server auth
Opens a browser for the OAuth flow; the token is saved to
~/Library/Application Support/whoop-mcp/token.json (macOS) and refreshed
automatically thereafter (rotating refresh tokens).
4. Register with Claude Code
claude mcp add whoop --scope user \
-e WHOOP_CLIENT_ID=... \
-e WHOOP_CLIENT_SECRET=... \
-- /path/to/whoop-mcp-server
Or for Claude Desktop, in claude_desktop_config.json:
{
"mcpServers": {
"whoop": {
"command": "/path/to/whoop-mcp-server",
"env": {
"WHOOP_CLIENT_ID": "...",
"WHOOP_CLIENT_SECRET": "..."
}
}
}
}
Example queries
- "What's my recovery score today?"
- "Show my sleep for the past week"
- "What's my average strain over the last 7 days?"
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.