Mcpmongo

🚀 MCP MongoDB Integration

This project demonstrates the integration of MongoDB with the Model Context Protocol (MCP), providing database interaction capabilities for AI assistants. The project consists of two main components:

1. MCP MongoDB Server - A server that offers MongoDB operations as MCP tools.
2. Client-side Gemini Integration - A terminal-based chatbot that uses Google's Gemini AI and accesses MongoDB tools.

📁 Project Structure

.
├── client-side/            # Client application using Gemini AI
│   ├── index.js            # Main client application
│   └── package.json        # Client dependencies
└── mcp-mongo-project/      # MCP server with MongoDB tools
    ├── src/
    │   ├── index.js        # Server setup and tool definitions
    │   └── services/
    │       └── mcp-service.js  # MongoDB service implementation
    └── package.json        # Server dependencies

✨ Features

• MongoDB Integration: Exposes full CRUD operations as MCP tools.
• MCP Server: Implements the Model Context Protocol for AI tool usage.
• Gemini AI Integration: Connects to Google's Gemini model.
• Terminal Chatbot: Provides an interactive chat interface for database operations.

📋 Prerequisites

• Node.js (v14 or higher)
• MongoDB instance (local or remote)
• Google Gemini API key

📦 Installation

1. Server Setup

# Navigate to the server directory
cd mcp-mongo-project

# Install dependencies
npm install

# Create a .env file containing MongoDB connection details
echo "MONGODB_URI=mongodb://localhost:27017" > .env
echo "MONGODB_NAME=your_database_name" >> .env

# Start the server
npm start

2. Client Setup

# Navigate to the client directory
cd client-side

# Install dependencies
npm install

# Create a .env file containing the Gemini API key
echo "GEMINI_API_KEY=your_gemini_api_key" > .env

# Start the client
node index.js

💻 Available MongoDB Tools

The MCP server exposes the following MongoDB operations as tools:

• findDocuments: Query documents in a collection.
• findOneDocument: Find a single document.
• insertOneDocument: Insert a single document.
• insertManyDocuments: Insert multiple documents.
• updateOneDocument: Update a single document.
• updateManyDocuments: Update multiple documents.
• deleteOneDocument: Delete a single document.
• deleteManyDocuments: Delete multiple documents.
• aggregateDocuments: Run an aggregation pipeline.
• countDocuments: Count the number of documents in a collection.
• listCollections: List all collections.
• createCollection: Create a new collection.

💻 Usage Examples

Once both the server and client are running, you can interact with MongoDB through the chat interface:

You: Show all collections in the database
AI: Found 3 collections in the database
[
  "users",
  "products",
  "orders"
]

You: Find all users older than 30
AI: Found 2 documents in the 'users' collection
[
  {
    "_id": "6450a7c63020e15b2a1a",
    "name": "Zhang San",
    "age": 35
  },
  {
    "_id": "6450a7d03020e15b2a1b",
    "name": "Li Si",
    "age": 32
  }
]

🔧 Technical Details

• MCP Protocol: Defines a standard interface for AI assistants to interact with external systems.
• MongoDB Integration: Connects the MongoDB database to Gemini AI via the MCP protocol.
• Terminal Chatbot: Provides a user-friendly interface for database queries and operations.

🖼️ Images

📄 License

[Insert license information here]

👥 Contributors

[List contributors here]

🔗 Links

[Add relevant links here]