trAIner MCP server
Commit Oct 26, 2026Provides AI-powered fitness coaching and personalized training plans by analyzing your Strava workout data.
Also known as Joe-Heffer/trainer
- Stars
- 3
- Tools
- —
- Downloads, 30 days
- —
- Latest version
- —
- Last commit
- Oct 26, 2026
- License
- MIT
What is trAIner MCP?
trAIner provides AI-powered fitness coaching and personalized training plans by analyzing your Strava workout data.
Summary generated from the project's README
About trAIner MCP server
Receiving personalized fitness coaching and customized training plans by connecting your workout history to an AI agent is what trainer enables. This tool acts as an agentic personal trainer that uses your Strava data to provide intelligent advice on training, recovery, and performance through a conversational interface.
To use this server, you need a Google Gemini API key and a Strava account. The setup requires a local installation of Python 3.10 or higher and a connection to the strava-mcp server. Once configured with your API key and the absolute path to the Strava MCP server, you can interact with the agent via the command line or through programmatic Python implementation using the TrainerAgent class.
Users can perform specific tasks such as analyzing a single activity via an activity_id or generating long-term training schedules. For example, you can request a plan to run a half marathon under two hours over a twelve-week period. The system uses your existing cycling distances and weight training frequency to suggest improvements like interval training or progressive overload.
What you can do with it
- 1Analyze specific workout performanceAsk the AI to provide detailed feedback on a specific past activity using its unique ID.
- 2Get performance improvement adviceChat with the agent to receive recommendations on how to improve specific disciplines like cycling.
Written from the project's README. Every use case is checked against a quote from it.
Configuration
GOOGLE_API_KEYLOG_LEVELSTRAVA_MCP_PATH
Automated checks
Ran Nov 14, 2026Full report →Not verified yet. A server needs a score of 80, no serious finding, and enough data for us to check. How checks work · What to fix for more points
- Source repository linked and reachablejoe-heffer/trainer is public and reachablegithub.com/joe-heffer/trainer ↗Pass
- Package names this repository as its sourceNot checked
- Build provenance (Sigstore attestation)Not checked
- Open source license (OSI)MITspdx.org/licenses/MIT.html ↗Pass
- Commit or release in the last yearLast commit or release 2026-10-26github.com/joe-heffer/trainer ↗Pass
- OpenSSF ScorecardNot checked
- Package name not close to a popular packageNot checked
- No critical advisory in the package or its dependenciesNot checked
- No hidden instructions to the model0 tool descriptions and the README read; no override phrases, credential paths or hidden charactersPass
- Destructive tools are labelledNot checked
- Remote endpoint answers the MCP handshakeNot checked
README
View on GitHubtrAIner
An open-source agentic AI personal trainer for fitness and health that's powered by Strava data.

Features
- AI-Powered Coaching: Get intelligent training advice powered by advanced LLMs
- Strava Integration: Automatically analyze your workouts from Strava
- Workout Analysis: Receive detailed feedback with personalized recommendations
- Training Plans: Generate customized plans based on your goals and fitness level
- Conversational Interface: Chat naturally with your AI trainer about training, recovery, and performance
Quick Start
Prerequisites
- Python 3.10+
- Google Gemini API key (get one here)
- Strava account (sign up)
Installation
-
Clone and install:
git clone https://github.com/Joe-Heffer/trainer.git cd trainer pip install -e . -
Set up Strava MCP Server:
# Clone and build the Strava MCP server git clone https://github.com/r-huijts/strava-mcp.git cd strava-mcp npm install npm run build # Authenticate with Strava (creates API app and gets tokens) npx tsx scripts/setup-auth.ts # Follow the prompts to authorize with Strava -
Configure environment:
cd ../trainer cp .env.example .env # Edit .env and add: # - GOOGLE_API_KEY: Your Google Gemini API key (https://aistudio.google.com/apikey) # - STRAVA_MCP_PATH: Absolute path to strava-mcp/dist/server.jsExample
.env:GOOGLE_API_KEY=your_gemini_api_key_here STRAVA_MCP_PATH=/absolute/path/to/strava-mcp/dist/server.js LOG_LEVEL=WARNING
Usage
Command line
trainer
For the arguments:
trainer --help
Programmatic
from trainer import TrainerAgent
agent = TrainerAgent()
await agent.initialize()
# Analyze a workout
analysis = await agent.analyze_workout("activity_id")
# Create a training plan
plan = await agent.create_training_plan(
goal="Run a half marathon under 2 hours",
weeks=12
)
Example Conversation
💪> What should I do to improve my cycling performance?
Based on your Strava data, here's what I see:
- You've been cycling consistently with multiple rides per week
- Your distances vary from 7km to 80km
- You're incorporating regular weight training which supports cycling performance
Recommendations:
1. **Structured Training Zones:** Add interval training to improve VO2 max
2. **Progressive Overload:** Gradually increase ride duration and intensity
3. **Strength Training:** Continue your weight training - it's great for power
4. **Recovery:** Ensure adequate rest between hard efforts
5. **Nutrition:** Focus on pre-ride and during-ride fueling for longer sessions
Would you like me to create a specific training plan for your cycling goals?
💪> quit
👋 Thanks for training with trAIner!
Documentation
For Users:
- Installation Guide - Detailed setup instructions
- MCP Integration - Connecting to Strava
For Contributors:
- Contributing Guide - How to contribute
- Developer Guide - Development setup and workflows
- Architecture - System design and components
- Testing Guide - Writing and running tests
Contributing
We welcome contributions! See CONTRIBUTING.md to get started.
Support
- Issues: GitHub Issues
- Discussions: GitHub Discussions
License
MIT License - see LICENSE for details.
Acknowledgments
Built with:
- Google Agent Development Kit (ADK)
- Model Context Protocol (MCP)
- r-huijts/strava-mcp - Strava MCP server
- Strava API
trAIner MCP questions
What data does the AI trainer use?
The trainer uses your Strava data to analyze workouts and provide intelligent coaching advice.