nutritionix

This MCP server exposes Nutritionix functionality to AI agents via the MCP protocol. It enables natural language tooling to search for foods (including common items and branded products), parse meals into nutritional breakdowns, and estimate calories burned for common exercises. Agents can invoke tools to perform quick food lookups, retrieve detailed nutrition data (calories, macros, vitamins, etc.), and get exercise calorie estimates, all through structured MCP calls. To use it, configure your MCP client (e.g., Claude Desktop) to point at the nutritionix-mcp server, then send tool invocations that request food searches, nutrition facts, or exercise calories. The server handles API requests to Nutritionix behind the scenes and returns structured results suitable for downstream reasoning and response formatting.

Prerequisites:

• Python 3.11 or newer
• Access to install local Python packages
• Nutritionix API credentials (App ID and App Key)

Installation steps:

1. Ensure Python is installed:

   • On macOS/Linux: sudo apt-get install python3.11 python3-venv
   • On Windows: install Python 3.11+ from the official installer

2. Install the Nutritionix MCP server package (the server executable should be exposed when the package is installed):

   • Install uv (Python package manager) if not already installed: python -m pip install uv
   • Install the Nutritionix MCP server package: python -m pip install nutritionix-mcp-server

3. Run the server (with your Nutritionix App ID and App Key): uvx nutritionix-mcp-server --app-id YOUR_APP_ID --app-key YOUR_APP_KEY

4. Optional: Save configuration for your MCP client (e.g., Claude Desktop) using the example in the README and replace placeholders with your actual credentials.

Prerequisites recap:

• Python 3.11+
• uv (or uvx-capable environment) installed
• Nutritionix API credentials
• Access to install Python packages

Notes and tips:

• You can pass APP_ID and APP_KEY as command-line arguments or environment variables (APP_ID, APP_KEY) depending on your deployment setup.
• If you rotate API credentials, update the running server configuration or environment accordingly.
• Ensure network access to Nutritionix API is allowed from the environment running the MCP server.
• In Claude or other MCP clients, use the provided server name (nutritionix-mcp) and supply app credentials as shown in the example config.
• If you run into authentication errors, double-check that the App ID and App Key are correct and have the necessary Nutritionix plan access for the endpoints you query.
• The server can be extended with additional tools (e.g., barcode search, food logging) by contributing to the MCP server project.