Overview
Strava MCP Server is a TypeScript-based MCP server that acts as a bridge between the Strava API and LLM-enabled workflows. It exposes Strava data and functionality as MCP tools, allowing large language models to fetch recent activities, athlete profiles, stats, and detailed activity streams. It supports segments, segment efforts, and route management, including listing, viewing, starring, and exporting routes as GPX or TCX files to the local filesystem. The server uses Strava API v3 and implements token management with automatic refreshing to maintain seamless access via environment variables. Setup includes installing dependencies, configuring Claude's MCP, and running an authentication flow to bind a Strava application (client id/secret) and store tokens in .env. The API reference outlines tools like get-recent-activities, get-athlete-profile, get-activity-details, list-athlete-clubs, get-activity-streams, and export-route-gpx/tcx. The server emphasizes AI-friendly JSON responses and robust error handling for missing tokens and API errors, along with optional route export path configuration.
Features
Access recent activities, profile, and stats
Fetches the authenticated user's recent activities, athlete profile, and statistics.
Fetch detailed activity streams
Retrieve time-series data streams (power, heart rate, cadence, etc.) for activities.
Explore, view, star, and manage segments
List, view, star/unstar, and manage Strava segments and landmarks.
View detailed activity and segment effort information
Provide in-depth details for activities and segment efforts.
List and view details of saved routes
List routes created by the athlete and show route details.
Export routes in GPX or TCX format
Export selected routes to GPX or TCX files saved locally.
AI-friendly JSON responses via MCP
Return responses in structured JSON tailored for LLM consumption.
Uses Strava API v3
Built on Strava API v3 for compatibility.
Who Is This For?
- LLM developers:Expose Strava data to LLM prompts via MCP tooling for automated analysis.
- Athletes:Query own Strava data via MCP tools for personal insights.
- AI copilots:Bridge Strava data into Claude workflows for task automation and insights.
