Mongo MCP Server

🚀 MongoDB MCP Server

This is a Model Context Protocol (MCP) server designed for MongoDB databases. The server can analyze the structure of MongoDB databases and use artificial intelligence to generate queries.

✨ Features

1. ERD Generation (generateErd)

Analyzes the collection structure of the database and generates an Entity-Relationship Diagram (ERD).

• Input Parameters:

{
    "database": "string",
    "collections": ["string"],  // Optional
    "format": "mermaid" | "json"  // Default: "mermaid"
}

• Usage Example:

{
    "database": "test",
    "format": "mermaid"
}

2. Query Generation (generateQuery)

Generates MongoDB queries based on natural language descriptions. Provides accurate queries by analyzing the actual structure of the database.

• Input Parameters:

{
    "database": "string",
    "collection": "string",
    "description": "string",
    "type": "mongodb" | "mongoose",  // Default: "mongodb"
    "includeExplanation": boolean  // Default: true
}

• Usage Example:

{
    "database": "test",
    "collection": "users",
    "description": "Find users who registered today",
    "type": "mongodb"
}

📦 Installation and Running

1. Install dependencies:

npm install

2. Build the project:

npm run build

3. Start the server:

node dist/index.js "mongodb://localhost:27017"

🛠️ Claude Desktop Setup

To use with Claude Desktop on macOS, follow these configuration steps:

1. Configuration file location: ~/Library/Application Support/Claude/claude_desktop_config.json

2. Add MCP server configuration:

{
    "mcp": {
        "servers": [
            {
                "name": "mongodb",
                "command": "node /path/to/mongo-mcp-server/dist/index.js mongodb://localhost:27017/test"
            }
        ]
    }
}

3. Restart Claude Desktop

🔧 Development

Project Structure

src/
├── index.ts          # Main server code
├── tools/
│   ├── generateErd.ts    # ERD generation tool
│   └── generateQuery.ts  # Query generation tool

Adding New Tools

1. Create a new tool file in the src/tools directory.
2. Implement the tool interface:

export const newTool = {
    parameters: {
        // Parameter definition
    },
    async execute(params: any, context: { client: MongoClient }) {
        // Tool logic implementation
    }
};

3. Register the new tool in src/index.ts.