Open Food Facts MCP Server

What is Open Food Facts MCP Server?

An MCP server implementation for Open Food Facts, a free, open, collaborative database of food products. It enables AI assistants to access food product information, offering nutritional analysis, product comparisons, and recipe suggestions. Designed for food-aware consumers who want to make informed dietary choices.

How to use Open Food Facts MCP Server?

Clone the repository, run npm install, then npm run build and npm start. Configure your AI assistant (e.g., Claude Desktop, VS Code Copilot) with the server’s command and transport mode (stdio or HTTP/SSE). For HTTP mode, set TRANSPORT=http PORT=3000 npm start and connect to http://localhost:3000/sse. Example conversation prompts are provided for searching, analyzing, comparing products, and checking allergens.

Key features of Open Food Facts MCP Server

• Search products by name, barcode, category, or brand
• Get AI-powered nutritional analysis and health scores
• Compare two products side‑by‑side with AI insights
• Check allergens and additives for dietary safety
• Suggest recipes using products you have
• Retrieve price history and current pricing data

Use cases of Open Food Facts MCP Server

• Quickly assess if a product is healthy by analyzing its nutrition and Nutri‑Score
• Find healthier alternatives by comparing two similar products
• Verify allergen safety before purchasing, e.g., “Does this product contain gluten?”
• Get recipe ideas using ingredients you already have at home
• Search for products by environmental impact (Eco‑Score) or processing level (NOVA)

FAQ from Open Food Facts MCP Server

What are the prerequisites for running the server?

Node.js v16 or higher and npm or yarn are required. The server runs on your local machine and communicates with the Open Food Facts API.

Which transports are supported?

The server supports stdio transport (for Claude Desktop, VS Code) and HTTP/SSE transport (for browser‑based AI tools). Set the TRANSPORT environment variable accordingly.

Why can’t my AI assistant find the tools?

Verify the MCP server is running and that your assistant’s configuration points to the correct command and transport type (stdio vs HTTP). Ensure the server’s path in the config matches the built dist/index.js file.

What if a product is not found in the database?

First use the searchProducts tool to locate the product. Confirm the barcode is 8–14 digits. Some products may not yet be indexed in Open Food Facts; you can contribute product data to the underlying database.

Where does the product data come from?

All data is fetched in real time from Open Food Facts, a free and open collaborative database of food products worldwide. No local storage is used; the server acts as a bridge to the public API.