Health & Fitness Coach

The Health & Fitness Coach MCP server provides a standardized interface that lets AI tools read, analyze, and act on your fitness data. It bridges natural language queries and structured data actions so you can log workouts, receive AI-generated plans, and get intelligent coaching across both web and AI interfaces.

How to use

You can connect your AI tools to the MCP server in two ways: via a remote HTTP endpoint or through a local command-line client. Once connected, you can ask your AI assistant to log workouts, view your progress, generate personalized plans, or set goals. The server translates your natural language requests into concrete data actions and returns results that the AI can use to craft coaching responses.

How to install

git clone https://github.com/your-username/health-fitness-coach-mcp.git
cd health-fitness-coach-mcp
npm install
cp env.example .env.local
# Add OPENAI_API_KEY for AI-generated plans
npm run dev

Configuration and usage notes

The MCP server exposes two connection methods you can use from your AI tools. The first is a remote HTTP interface, typically available at a local development URL like http://localhost:3000/sse. The second is a local STDIO setup that runs a command such as npx -y mcp-remote http://localhost:3000/mcp to connect a desktop AI client directly to the server. You can use either method depending on your tooling preference.

Real-world usage scenarios

Daily Workout Logging: Tell your AI assistant you completed a workout; the MCP server logs the session and updates progress while the web dashboard reflects the new totals.

Intelligent Plan Generation: Ask for a workout plan based on your week’s activity; the AI analyzes your context and produces a tailored routine and meals suggestions.

Security and access considerations

Keep your API keys and tokens secure. Limit access to the MCP server endpoints and use proper authentication when exposing the HTTP interface in production. Review environment variable management and rotate keys periodically to minimize risk.

Deployment options

Local development runs the MCP server on http://localhost:3000/mcp with the web interface at http://localhost:3000. For production, deploy to your hosting provider of choice and configure the environment with your AI API keys and any caching or database connections you need.

Available tools

log-workout

Records workout sessions with intelligent categorization and stores details like userId, date, type, duration, and distance.

log-nutrition

Logs meals with nutritional analysis, capturing userId, date, meal type, items, and calories.

log-feedback

Captures subjective progress notes and motivation data for ongoing coaching.

generate-plan

Creates personalized workout and nutrition plans based on history, goals, and preferences.

view-context

Returns a comprehensive fitness profile including history, patterns, goals, and insights for AI decision-making.

set-weekly-target

Sets and tracks weekly fitness goals such as cardio targets and calorie budgets.

echo

System health check tool to verify MCP connectivity and responsiveness.