# TrainingPeaks MCP Server

<a href="https://glama.ai/mcp/servers/@tildecomunicacion/TrainingPeaks-MCP">
  <img width="380" height="200" src="https://glama.ai/mcp/servers/@tildecomunicacion/TrainingPeaks-MCP/badge" alt="TrainingPeaks MCP server" />
</a>

Connect TrainingPeaks to Claude and other AI assistants via the Model Context Protocol (MCP). Query workouts, build structured intervals, track fitness trends, and write comments/feedback on your calendar through natural conversation.

**No more expired cookies.** This server features **automatic login** using your TrainingPeaks username and password. The credentials remain local on your machine and are only transmitted directly to TrainingPeaks over secure HTTPS.

---

## What You Can Do

Ask your AI assistant things like:
- "Build me a 4x8min threshold session for Tuesday with warm-up and cool-down"
- "Show my workouts for this week and check if I've completed them"
- "Analyze my compliance for yesterday's run and add a feedback comment in TrainingPeaks"
- "What's my CTL, ATL, and TSB trend for the last 90 days?"
- "Get my profile and account type"

---

## Tools (8)

### Workouts & Intervals
- **`tp_get_workouts`**: List planned and completed workouts in a date range (max 90 days).
- **`tp_create_workout`**: Create planned workouts on a calendar date (supports complex nested interval structures, auto-computed TSS/IF, and planned start times).
- **`tp_delete_workout`**: Delete planned workouts by ID.

### Comments & Feedback
- **`tp_get_workout_comments`**: Fetch the list of comments from a workout.
- **`tp_add_workout_comment`**: Add a text comment (e.g. analysis, compliance review) to a workout.

### Performance & Summary
- **`tp_get_profile`**: Get athlete profile info (Athlete ID, Name, Email, Premium/Basic account status).
- **`tp_get_weekly_summary`**: Aggregate view of workouts, total TSS, total duration, and end-of-week fitness.
- **`tp_get_fitness`**: Fetch CTL (Fitness), ATL (Fatigue), and TSB (Form) performance metrics.

---

## Setup in Claude Desktop

Add the server to your `claude_desktop_config.json`:
- **Windows:** `%APPDATA%\Claude\claude_desktop_config.json`
- **macOS:** `~/Library/Application Support/Claude/claude_desktop_config.json`

```json
{
  "mcpServers": {
    "trainingpeaks": {
      "command": "python",
      "args": [
        "-m",
        "tp_mcp",
        "serve"
      ],
      "env": {
        "TP_USERNAME": "YOUR_TRAININGPEAKS_USERNAME",
        "TP_PASSWORD": "YOUR_TRAININGPEAKS_PASSWORD"
      }
    }
  }
}
```

> [!NOTE]
> Ensure you run this inside the virtual environment where `tp-mcp` is installed, or point the `"command"` path directly to the `python.exe`/`python` binary of the virtual environment.

---

## Security & Privacy

This server is designed to be completely secure and private:
1. **100% Local:** Runs on your local machine. No external databases, no middleware, no credential sharing.
2. **Direct Connection:** Authenticates directly with TrainingPeaks official servers using HTTPS.
3. **No exposed ports:** Uses standard input/output (stdio) to communicate with Claude Desktop. It does not open any network ports.

---

## Development & Testing

To set up the project locally and run the tests:

```bash
# Clone the repository
git clone https://github.com/tildecomunicacion/trainingpeaks_mcp.git
cd trainingpeaks-mcp

# Create a virtual environment and install dependencies
python -m venv .venv
source .venv/bin/activate  # Windows: .venv\Scripts\activate
pip install -e ".[dev]"

# Run unit tests
pytest
```

---

## License

MIT