strava-mcp
Exposes your Strava data to Claude, enabling training load analysis, activity comparison, and lap splits via conversation.
README
strava-mcp
A Model Context Protocol (MCP) server that exposes your Strava data to Claude. Ask Claude about your training load, compare blocks, dig into lap splits, or analyze activity streams — all from a conversation.
Prerequisites
- Node.js 18+
- A Strava account with activities
- A Strava API application (free to create)
Strava API setup
- Go to strava.com/settings/api and create an application.
- Note your Client ID and Client Secret.
- Obtain an access token and refresh token using Strava's OAuth flow. The simplest way:
- Visit
https://www.strava.com/oauth/authorize?client_id=YOUR_CLIENT_ID&response_type=code&redirect_uri=http://localhost&approval_prompt=force&scope=read,activity:read_all - Authorize the app and copy the
codefrom the redirect URL. - Exchange it for tokens:
curl -X POST https://www.strava.com/oauth/token \ -d client_id=YOUR_CLIENT_ID \ -d client_secret=YOUR_CLIENT_SECRET \ -d code=YOUR_CODE \ -d grant_type=authorization_code - The response contains
access_token,refresh_token, andexpires_at.
- Visit
Installation
git clone https://github.com/your-username/strava-mcp.git
cd strava-mcp
npm install
Configuration
cp .env.example .env
Edit .env with your credentials:
STRAVA_CLIENT_ID=your_client_id
STRAVA_CLIENT_SECRET=your_client_secret
STRAVA_ACCESS_TOKEN=your_access_token
STRAVA_REFRESH_TOKEN=your_refresh_token
STRAVA_TOKEN_EXPIRES_AT=1234567890 # unix timestamp from token exchange
DEBUG=false # set to "true" for verbose stderr logging
Tokens are held in memory only and auto-refreshed before expiry. Nothing is written back to .env.
Build & run
npm run build # compile TypeScript → build/
npm start # start the MCP server on stdio
For development without a build step:
npm run dev
Claude Desktop integration
Add this server to your Claude Desktop config (~/Library/Application Support/Claude/claude_desktop_config.json on macOS):
{
"mcpServers": {
"strava": {
"command": "node",
"args": ["/absolute/path/to/strava-mcp/build/index.js"],
"env": {
"STRAVA_CLIENT_ID": "your_client_id",
"STRAVA_CLIENT_SECRET": "your_client_secret",
"STRAVA_ACCESS_TOKEN": "your_access_token",
"STRAVA_REFRESH_TOKEN": "your_refresh_token",
"STRAVA_TOKEN_EXPIRES_AT": "0"
}
}
}
}
Restart Claude Desktop after saving.
Available tools
Athlete
| Tool | Description |
|---|---|
get_athlete |
Profile: name, location, weight, FTP, membership type |
get_athlete_stats |
Activity totals (recent / YTD / all-time) for rides, runs, swims |
get_athlete_zones |
Heart rate and power training zones from your Strava settings |
Activities
| Tool | Description |
|---|---|
list_activities |
Recent activities with optional date range and count cap |
get_activity |
Full detail for a single activity by ID |
get_activity_laps |
Lap splits (pace, HR, power) for a single activity |
get_activity_zones |
HR and power zone distributions for a single activity (Strava Summit required) |
search_activities |
Filter recent activities by name keyword and/or sport type |
Analysis
| Tool | Description |
|---|---|
analyze_training_load |
Volume, time, elevation, and weekly breakdown over a date range |
compare_training_blocks |
Side-by-side comparison of two training periods with deltas |
get_training_timeseries |
Week-by-week or day-by-day volume trend over a date range |
get_activity_streams |
Raw sensor data (HR, pace, power, altitude, cadence) with summary stats |
Example prompts
- "How many miles did I run last month?"
- "Compare my training in January vs February."
- "Show me the lap splits for my last long run."
- "What was my average heart rate trend over the past 12 weeks?"
- "Find all my rides over 50 miles this year."
Project structure
src/
index.ts — MCP server entry point
types.ts — TypeScript interfaces and error classes
utils.ts — Unit conversion helpers (imperial: miles, min/mile, feet)
logger.ts — Debug logger (writes to stderr only)
token-manager.ts — In-memory OAuth token lifecycle (refresh, expiry)
strava-client.ts — Authenticated HTTP client with pagination
tools/
athlete.ts — Athlete profile and stats tools
activities.ts — Activity listing and detail tools
analysis.ts — Training load and trend analysis tools
Notes
- Units: All display values are imperial (miles, min/mile, feet). Strava's metric API values are converted via
utils.ts. - Rate limits: Strava enforces 100 requests/15 min and 1,000/day. Tools that fetch large date ranges may hit these limits.
- Pagination: The client handles pagination automatically; pass
maxItemsto cap results. - Logging: Debug output goes to stderr only, so it never corrupts the MCP stdout stream. Enable with
DEBUG=true.
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.