ogerbron/trainingpeaks-mcp-server

TrainingPeaks MCP Server

An MCP (Model Context Protocol) server that enables Claude to interact with the TrainingPeaks API for retrieving athlete data, workouts, metrics, and calendar events.

Features

• Athlete Profile: Get basic profile information and training zones
• Workouts: Retrieve workout history and detailed workout information
• Calendar Events: Access TrainingPeaks calendar events
• Metrics: Get health and fitness metrics (weight, HRV, steps, stress, sleep)
• Planned Workouts: Retrieve upcoming planned workouts (up to 7 days)
• OAuth 2.0 Authentication: Secure authentication with TrainingPeaks API

Prerequisites

• Python 3.8+
• TrainingPeaks API credentials (client_id, client_secret)
• API access approval from TrainingPeaks (required for production use)

Installation

1. Clone the repository:

git clone <repository-url>
cd trainingpeaks-mcp-server

2. Install dependencies:

pip install -e .

3. Copy environment configuration:

cp .env.example .env

4. Edit .env with your TrainingPeaks API credentials:

TRAININGPEAKS_CLIENT_ID=your_client_id_here
TRAININGPEAKS_CLIENT_SECRET=your_client_secret_here
TRAININGPEAKS_REDIRECT_URI=http://localhost:8080/callback
TRAININGPEAKS_SCOPES=athlete:profile,athlete:workouts
TRAININGPEAKS_ENVIRONMENT=sandbox

Usage

Running the MCP Server

trainingpeaks-mcp-server

Or alternatively:

python -m trainingpeaks_mcp_server.server

To test that the server works:

python test_server.py

Available Tools

1. get_athlete_profile: Get athlete's profile and training zones
2. get_workouts: Retrieve workouts with optional date filtering
3. get_workout_details: Get detailed information about a specific workout
4. get_calendar_events: Access calendar events within a date range
5. get_metrics: Get health metrics (weight, HRV, steps, stress, sleep)
6. get_planned_workouts: Retrieve upcoming planned workouts
7. set_auth_tokens: Set OAuth tokens for API authentication

Authentication

Before using the API tools, you need to authenticate:

1. Use TrainingPeaks OAuth flow to obtain access and refresh tokens
2. Use the set_auth_tokens tool to provide tokens to the MCP server
3. The server will automatically handle token refresh when needed

API Access

Important: TrainingPeaks API access is currently limited to approved commercial applications. To request access:

1. Visit https://api.trainingpeaks.com/request-access
2. Provide information about your application and use case
3. Allow 7-10 days for response

For development/testing, use the sandbox environment.

Configuration

The server supports both production and sandbox environments:

• Sandbox: TRAININGPEAKS_ENVIRONMENT=sandbox
• Production: TRAININGPEAKS_ENVIRONMENT=production

Plug Claude

Set the following in your claude configuration:

{
  "mcpServers": {
    "trainingpeaks": {
      "command": "trainingpeaks-mcp-server",
      "env": {
        "TRAININGPEAKS_CLIENT_ID": "your_client_id_here",
        "TRAININGPEAKS_CLIENT_SECRET": "your_client_secret_here",
        "TRAININGPEAKS_ENVIRONMENT": "sandbox"
      }
    }
  }
}

See .

Development

Install development dependencies:

pip install -e ".[dev]"

Run tests:

pytest

Format code:

black src/
ruff check src/

License

MIT License - see LICENSE file for details.