FitBit MCP Server

What is Fitbit MCP?

Fitbit MCP (Model Context Protocol) is an implementation that enables AI assistants to access and analyze your Fitbit health and fitness data. It acts as an MCP server that connects to the Fitbit API using a token obtained from an OAuth flow.

How to use Fitbit MCP?

Fitbit MCP can be used with AI assistant frameworks by configuring it with a JSON object. You can provide the Fitbit access token either as an environment variable (FITBIT_ACCESS_TOKEN) or as a command-line argument (--fitbit-token).

Example JSON configuration:

{
  "command": "npx",
  "args": ["-y", "fitbit-mcp", "--stdio"],
  "env": {
    "FITBIT_ACCESS_TOKEN": "YOUR_FITBIT_ACCESS_TOKEN"
  }
}

To obtain a Fitbit access token, you need to create an application on the Fitbit Developer Portal, set the OAuth 2.0 Application Type to "Personal", and the Callback URL to "http://localhost:3000". Then, use the OAuth 2.0 authorization flow to get the token.

Key Features of Fitbit MCP

Fitbit MCP provides a variety of tools for AI assistants to access different types of Fitbit data:

getUserProfile: Get your Fitbit profile information.
getActivities: Get activity data for a specified date.
getSleepLogs: Get sleep data for a specified date.
getHeartRate: Get heart rate data for a specified date and period.
getSteps: Get step count for a specified date and period.
getBodyMeasurements: Get weight and body fat data.
getFoodLogs: Get food log data for a specified date.
getWaterLogs: Get water consumption data for a specified date.
getLifetimeStats: Get lifetime activity statistics.
getUserSettings: Get user settings and preferences.
getFloorsClimbed: Get floors climbed data.
getDistance: Get distance data.
getCalories: Get calories burned data.
getActiveZoneMinutes: Get active zone minutes data.
getDevices: Get information about connected Fitbit devices.
getBadges: Get earned badges and achievements.

Most tools accept optional date (YYYY-MM-DD) and period (1d, 7d, 30d, 1w, 1m) parameters.

Use Cases of Fitbit MCP

Personalized Health Dashboards: AI assistants can fetch and display a user's health metrics from Fitbit in a customized dashboard.
Fitness Tracking and Analysis: Analyze activity, sleep, and heart rate data to provide insights into fitness trends.
Dietary Monitoring: Track food and water intake for health and wellness applications.
Automated Health Reports: Generate regular reports on a user's health and fitness progress.
Integration with AI-powered Health Coaches: Allow AI coaches to access real-time Fitbit data to provide personalized recommendations.

FAQ about Fitbit MCP

Q: How do I get a Fitbit access token? A: You need to create an application on the Fitbit Developer Portal, set the OAuth 2.0 Application Type to "Personal", and the Callback URL to "http://localhost:3000". Then, follow the OAuth 2.0 authorization flow to obtain the token.

Q: What data can I access with Fitbit MCP? A: You can access a wide range of data including user profile, activities, sleep logs, heart rate, steps, body measurements, food logs, water logs, lifetime stats, user settings, floors climbed, distance, calories, active zone minutes, connected devices, and earned badges.

Fitbit MCP (Model Context Protocol)

A Model Context Protocol (MCP) implementation for Fitbit, enabling AI assistants to access and analyze your Fitbit health and fitness data.

Usage

For JSON configuration (for use with AI assistant frameworks):

{
  "command": "npx",
  "args": ["-y", "fitbit-mcp", "--stdio"],
  "env": {
    "FITBIT_ACCESS_TOKEN": "YOUR_FITBIT_ACCESS_TOKEN"
  }
}

Or with arguments instead of environment variables:

{
  "command": "npx",
  "args": ["-y", "fitbit-mcp", "--stdio", "--fitbit-token=YOUR_FITBIT_ACCESS_TOKEN"]
}

Available Tools

This MCP provides the following tools for AI assistants to access your Fitbit data:

getUserProfile: Get your Fitbit profile information
getActivities: Get activity data for a specified date
getSleepLogs: Get sleep data for a specified date
getHeartRate: Get heart rate data for a specified date and period
getSteps: Get step count for a specified date and period
getBodyMeasurements: Get weight and body fat data
getFoodLogs: Get food log data for a specified date
getWaterLogs: Get water consumption data for a specified date
getLifetimeStats: Get lifetime activity statistics
getUserSettings: Get user settings and preferences
getFloorsClimbed: Get floors climbed data
getDistance: Get distance data
getCalories: Get calories burned data
getActiveZoneMinutes: Get active zone minutes data
getDevices: Get information about connected Fitbit devices
getBadges: Get earned badges and achievements

Most tools accept optional parameters:

date: Date in YYYY-MM-DD format (defaults to today)
period: Time period for data (1d, 7d, 30d, 1w, 1m)

Obtaining a Fitbit Access Token

To get a Fitbit access token:

Create an application at Fitbit Developer Portal
Set OAuth 2.0 Application Type to "Personal"
Set Callback URL to "http://localhost:3000"
After creating the application, note your Client ID and Client Secret
Use the OAuth 2.0 authorization flow to obtain an access token

For detailed instructions on OAuth authentication, see the Fitbit API Documentation.

Contributing

Contributions are welcome! Here's how you can contribute:

Fork the repository
Create your feature branch (git checkout -b feature/amazing-feature)
Commit your changes (git commit -m 'Add some amazing feature')
Push to the branch (git push origin feature/amazing-feature)
Open a Pull Request

Every pull request triggers a GitHub Actions workflow that verifies the build process.

Development Setup

# Clone the repository
git clone https://github.com/your-username/fitbit-mcp.git
cd fitbit-mcp

# Install dependencies
npm install

# Build the project
npm run build

# Run in development mode
npm run dev

Release Process

To publish a new version to NPM:

Update the version in package.json
Create a new GitHub release with a tag like v1.0.1
The GitHub Actions workflow will automatically build and publish the package to NPM

Make sure you have the NPM_TOKEN secret configured in your GitHub repository settings.

License

This project is licensed under the MIT License - see the LICENSE file for details.