Power your agent
पब्लिशर का इंस्टॉल। PlugMCP एंडपॉइंट नहीं।
Host a copy with our keys and logs.
रजिस्ट्री में अभी कोई पैकेज्ड रिलीज़ नहीं है। रिपॉज़िटरी का README देखें.
रिमोट URL या आपकी मशीन पर कमांड? दोनों इंस्टॉल में क्या अंतर है
PlugMCP पर चलाएँ
A new server you create. It is not this listing's official endpoint. Use the publisher install when you want their URL and their credentials.
लॉन्चA MCP (Model Context Protocol) server that provides access to the Open Food Facts dataset using DuckDB and parquet for fast queries. Supports both local Claude Desktop integration and remote deployment with authentication.

This MCP server can operate in two distinct modes:
./openfoodfacts-mcp-server --stdio./openfoodfacts-mcp-server (default mode)/health endpoint)This MCP server downloads and caches the Open Food Facts Parquet dataset locally, then uses DuckDB for fast product searches. It provides two main tools:
The server automatically manages dataset updates, uses file locking for concurrent safety, and provides structured JSON logging.
This setup uses STDIO mode for local Claude Desktop integration.
script/build --simple
openfoodfacts-mcp-server --fetch-db
Add this to your Claude Desktop MCP settings (~/Library/Application Support/Claude/claude_desktop_config.json on macOS):
{
"mcpServers": {
"openfoodfacts": {
"command": "/path/to/openfoodfacts-mcp-server",
"args": ["--stdio"],
"env": {
"OPENFOODFACTS_MCP_TOKEN": "your-secret-token",
"DATA_DIR": "/full/path/to/openfoodfacts-mcp-server/data",
"ENV": "development"
}
}
}
}
Restart Claude Desktop. The mcp server will automatically start and be ready for food product queries.
This setup uses HTTP mode for remote deployment with authentication.
For production deployment (HTTP mode), configure these environment variables:
# Required: Authentication
OPENFOODFACTS_MCP_TOKEN=your-production-secret-token
# Optional: Data management
DATA_DIR=./data
PARQUET_URL=https://huggingface.co/datasets/openfoodfacts/product-database/resolve/main/product-database.parquet
REFRESH_INTERVAL_SECONDS=86400
# Optional: Server configuration
PORT=8080
ENV=production
# Optional: DuckDB Performance Tuning
DUCKDB_MEMORY_LIMIT=4GB # Memory limit (e.g. 2GB, 4GB, 8GB, 16GB)
DUCKDB_THREADS=4 # Number of threads (1-16 typically)
DUCKDB_CHECKPOINT_THRESHOLD=1GB # When to write to disk (e.g. 512MB, 1GB, 2GB)
DUCKDB_PRESERVE_INSERTION_ORDER=true # Set to false for large datasets to reduce memory
For remote deployment, run without the --stdio flag (HTTP mode is the default):
./openfoodfacts-mcp-server
This will start an HTTP server on the configured port (default 8080) with:
/health endpoint (no authentication required)/mcp endpoint (Bearer token authentication required)A cool tip for developing locally, you can actually do this and it will return a result from the MCP server:
echo '{"jsonrpc": "2.0", "id": 1, "method": "tools/call", "params": {"name": "search_products_by_brand_and_name_simplified", "arguments": {"name": "cream soda", "brand": "olipop", "limit": 1}}}' | go run cmd/openfoodfacts-mcp-server/main.go --stdio
MCP Server for OpenFoodFacts using the Parquet product database
टूल
अभी टूल सूची नहीं है। रिमोट सर्वर को पहले हेल्थ चेक के बाद सूची मिलती है।
FAQ
इस पेज पर Cursor, Claude Desktop, Claude Code या VS Code के लिए stdio कॉन्फ़िग कॉपी करें। यह पब्लिशर का इंस्टॉल इस्तेमाल करता है, PlugMCP प्रॉक्सी नहीं।
इस रिकॉर्ड में कोई ज़रूरी एनवायरनमेंट वेरिएबल दर्ज नहीं है।
Openfoodfacts stdio पर लोकल चलता है।
पिछली जाँच 2026-10-28 को हुई। MCP Server for OpenFoodFacts using the Parquet product database