USDA api MCP Server
Este servidor permite que você faça perguntas com informações nutricionais muito mais precisas.
Documentação
🍎 USDA Food Tools for Claude
Add powerful food and nutrition data to Claude for Desktop using the USDA's comprehensive food database.
🚀 For Users - Quick Install
Option 1: GUI Installer (Recommended)
- Download:
USDA Food Tools Installer.zip(11KB) - Extract and double-click the app
- Get API Key: Free at api.nal.usda.gov
- Follow the installer - opens in your browser
- Restart Claude - tools appear automatically!
Option 2: Command Line
curl -sSL https://raw.githubusercontent.com/yourusername/usda-api-mcp/main/install.sh | bash
🔍 What You Get
- Search Foods - Find any food with detailed nutrition facts
- Nutrition Analysis - Complete nutrient breakdowns
- Brand Information - Ingredients, categories, manufacturers
- Bulk Lookups - Multiple foods at once
- Research Data - USDA's gold-standard food database
📋 Requirements
- Claude for Desktop (latest version)
- Free USDA API key from api.nal.usda.gov
- macOS 10.15+ (GUI installer) or Python 3.11+ (command line)
💬 Example Usage
Once installed, ask Claude:
- "What's the nutrition info for salmon?"
- "Compare chicken breast vs tofu protein"
- "Find high-fiber breakfast foods"
- "Show vitamin content of spinach"
🛠️ For Developers
Development Setup
# Clone the repository
git clone https://github.com/yourusername/usda-api-mcp.git
cd usda-api-mcp
# Install dependencies
uv sync
# Set up environment
echo "USDA_API_KEY=your_key_here" > .env
# Run the MCP server
uv run main.py
Project Structure
usda-api-mcp/
├── main.py # MCP server implementation
├── gui_installer.py # Web-based GUI installer
├── install.sh # Command-line installer
├── pyproject.toml # Dependencies
├── create_app.sh # Build Mac app bundle
└── dist/ # Distribution files
Available MCP Tools
search_foods(query, page_size, page_number)- Search food databaseget_food_details(fdc_id, nutrients)- Get detailed food informationget_multiple_foods(fdc_ids, nutrients)- Bulk food lookuplist_foods(data_type, page_size, page_number)- Browse foodsget_food_nutrients(fdc_ids, nutrients)- Get specific nutrients
Building Releases
# Build Mac app bundle
./create_app.sh
# Create distribution zip
cd dist && zip -r "USDA-Food-Tools-Installer.zip" "USDA Food Tools Installer.app"
# Test the app
open "USDA Food Tools Installer.app"
API Integration
The server integrates with USDA FoodData Central API v2:
- Base URL:
https://api.nal.usda.gov/fdc/v2/ - Authentication: API key in query parameters
- Rate Limits: 3600 requests/hour (free tier)
- Data Types: Foundation, SR Legacy, Survey, Branded foods
Contributing
- Fork the repository
- Create a feature branch:
git checkout -b feature-name - Make your changes and test thoroughly
- Submit a pull request
Testing
# Test MCP server directly
uv run main.py
# Test GUI installer
python3 gui_installer.py
# Test command-line installer
./install.sh
Deployment
The project uses a simple Mac app bundle approach:
- No complex packaging (py2app issues resolved)
- Shell script launcher with embedded Python
- Self-contained with port conflict resolution
- Automatic cleanup of existing processes