ExampleApp MCP Server

A Model Context Protocol (MCP) server that exposes the GenericSuite ExampleApp food and nutrition management tools to AI clients like Claude Desktop, VS Code, and other MCP-compatible applications.

Features

This MCP server provides comprehensive food and nutrition management capabilities:

🔍 FDA Food Database Integration

get_fda_food_query: Search FDA food database for nutritional information

🥘 User Ingredient Management

create_user_ingredient: Add new ingredients to personal list
search_user_ingredient: Search ingredients by name
get_user_ingredient_list: Get all saved ingredients

🍽️ Dish Management

create_dish: Create dishes with multiple ingredients
search_dishes: Search dishes by name
get_dishes_list: Get all created dishes

📅 Daily Meal Tracking

create_daily_meal: Add daily meal entries
search_daily_meals: Search meals by date
get_daily_meal_list: Get meal history

👤 User Profile & Health

get_user_profile: Get basic health profile
get_full_user_profile: Get complete profile with eating habits

📂 Resources

user://profile: User profile data
user://ingredients: User ingredients data
user://dishes: User dishes data
user://meals: User meals data

📝 Prompts

nutrition_analysis: Generate nutrition analysis
meal_planning: Generate meal planning advice

Installation

Install Python dependencies: bash cd site/Sample-Code/exampleapp/apps/mcp-server pip install -r requirements.txt
Run the server: bash python mcp_server.py

MCP Client Configuration

Claude Desktop

Add this configuration to your claude_desktop_config.json:

{
  "mcpServers": {
    "exampleapp": {
      "command": "sh",
      "args": [
        "/absolute/path/to/genericsuite-basecamp/docs/Sample-Code/exampleapp/apps/mcp-server/run_mcp_server.sh"
      ],
      "env": {
        "GS_USER_ID": "xxxx",
        "GS_API_KEY": "xxxx"
      }
    }
  }
}

VS Code MCP Extension

Add to your VS Code settings or .vscode/mcp.json:

{
  "mcp": {
    "servers": {
      "exampleapp": {
        "command": "sh",
        "args": [
          "/absolute/path/to/genericsuite-basecamp/docs/Sample-Code/exampleapp/apps/mcp-server/run_mcp_server.sh"
        ],
        "env": {
          "GS_USER_ID": "xxxx",
          "GS_API_KEY": "xxxx"
        }
      }
    }
  }
}

Usage Examples

Once connected to an MCP client, you can interact with the server using natural language:

Basic Operations

"Search for apple in the FDA database"
"Add quinoa to my ingredients list with 368 kcal per 100g"
"Create a new dish called 'Healthy Breakfast' with Greek yogurt and berries"

Daily Meal Tracking

"Add a breakfast meal for today with 2 eggs and toast"
"What did I eat yesterday?"
"Show me my total calories for 2024-01-15"

Profile and Analysis

"Show me my nutrition profile"
"Analyze my eating patterns from last week"
"Give me meal planning suggestions for weight loss"

Architecture

The server is built using: - FastMCP: Modern Python MCP framework for rapid development - Pydantic: Data validation and serialization - AsyncIO: Asynchronous operations for better performance

Data Models

Comprehensive validation for all nutritional data
Support for various measurement units (grams, ounces, cups, etc.)
Automatic calorie calculations for recipes and meals

Mock Data

The server includes sample data for testing and demonstration: - Sample ingredients (quinoa, Greek yogurt, almonds) - Sample dishes (yogurt with almonds) - FDA food database entries (apple, chicken breast, rice)

Development

Extending the Server

To add new tools, create functions decorated with @mcp.tool():

@mcp.tool()
async def my_new_tool(params: MyModel) -> Dict[str, Any]:
    """Description of what this tool does"""
    # Implementation
    return {"success": True, "data": result}

Adding Resources

To add new resources, use the @mcp.resource() decorator:

@mcp.resource("mydata://endpoint")
async def my_resource() -> str:
    """Get my data as a resource"""
    data = await get_my_data()
    return json.dumps(data, indent=2)

Integration with GenericSuite's ExampleApp

This MCP server is based on the ExampleApp's AI tools: - ai_gpt_fn_app.py: Core food and nutrition tools - ai_gpt_fn_fda.py: Food database integrations

It provides the same functionality but exposed through the standardized MCP protocol, making it accessible to any MCP-compatible AI client.

Transport Modes

Currently supports: - STDIO: Standard input/output for local clients (default) - Future: HTTP/SSE transport for remote clients (planned)

Logging

The server provides detailed logging for debugging and monitoring: - Tool invocations and parameters - Data validation errors - Mock database operations - Performance metrics

License

This MCP server follows the same license as the GenericSuite project.