Hevy Fitness MCP Server

A Model Context Protocol (MCP) server that provides AI assistants with access to the Hevy fitness tracking API. This allows you to log workouts, manage routines, browse exercises, and track your fitness progress directly through AI chat interfaces.

🏋️ Features

This MCP server provides comprehensive access to Hevy's fitness tracking capabilities:

Workouts

• get_workouts - Browse your workout history (paginated)
• get_workout - Get detailed information about a specific workout
• create_workout - Log a new workout with exercises, sets, weights, and reps
• update_workout - Update an existing workout
• get_workouts_count - Get total number of workouts logged
• get_workout_events - Get workout change events (updates/deletes) since a date for syncing

Routines

• get_routines - List your workout routines
• get_routine - Get details of a specific routine
• create_routine - Create a new workout routine template
• update_routine - Update an existing routine

Exercises

• get_exercise_templates - Browse available exercises (includes both Hevy's library and your custom exercises)
• get_exercise_template - Get detailed information about a specific exercise template
• create_exercise_template - Create a custom exercise template
• get_exercise_history - View your performance history for a specific exercise

Organization

• get_routine_folders - List your routine folders for organization
• get_routine_folder - Get details of a specific routine folder
• create_routine_folder - Create a new routine folder

🚀 Quick Start

Prerequisites

1. Hevy Pro subscription - The Hevy API is only available to Pro users
2. Hevy API Key - Get yours at https://hevy.com/settings?developer
3. Cloudflare account - For deploying the MCP server

Deploy to Cloudflare Workers

1. Clone this repository:

git clone https://github.com/tomtorggler/hevy-mcp-server.git
cd hevy-mcp-server

2. Install dependencies:

npm install

3. Set your Hevy API key as a secret:

npx wrangler secret put HEVY_API_KEY
# Paste your API key when prompted

4. Deploy to Cloudflare:

npm run deploy

Your MCP server will be available at: https://hevy-mcp-server.<your-account>.workers.dev/mcp

Local Development

Run the server locally:

npm run dev

The server will be available at: http://localhost:8787/mcp

🔌 Connect to AI Clients

Claude Desktop

To connect from Claude Desktop, edit your config file (Settings > Developer > Edit Config):

{
  "mcpServers": {
    "hevy": {
      "command": "npx",
      "args": [
        "mcp-remote",
        "https://hevy-mcp-server.<your-account>.workers.dev/mcp"
      ]
    }
  }
}

Restart Claude Desktop and you'll see the Hevy tools available.

Cloudflare AI Playground

1. Go to https://playground.ai.cloudflare.com/
2. Enter your deployed MCP server URL
3. Start using Hevy tools directly from the playground!

📖 Usage Examples

Creating a Workout

Once connected, you can ask your AI assistant to log workouts:

"Log a workout from today at 10am to 11am. I did bench press: 3 sets of 100kg for 10 reps, and squats: 4 sets of 120kg for 8 reps."

The assistant will:

1. Use get_exercise_templates to find the exercise IDs
2. Call create_workout with the proper structure
3. Confirm the workout was logged successfully

Viewing Progress

"Show me my last 5 workouts"

"What's my exercise history for deadlifts?"

"Get all workout changes since January 1st, 2024"

The assistant will use get_workout_events to sync recent changes.

Managing Routines

"Create a new Push Day routine with bench press (4 sets of 8-12 reps at 100kg) and overhead press (3 sets of 10 rep

... View full README on GitHub