Mongo MCP

🚀 MongoDB MCP Server for LLM

A Model Context Protocol (MCP) server that enables Large Language Models (LLMs) to interact directly with MongoDB databases. Query collections, check schemas, analyze data, and perform management tasks using natural language.

🚀 Quick Start

To get started, find your MongoDB connection URL and add the following configuration to your Claude Desktop configuration file:

MacOS: ~/Library/Application\ Support/Claude/claude_desktop_config.json
Windows: %APPDATA%/Claude/claude_desktop_config.json

{
  "mcpServers": {
    "mongodb": {
      "command": "npx",
      "args": [
        "mongo-mcp",
        "mongodb://<username>:<password>@<host>:<port>/<database>?authSource=admin"
      ]
    }
  }
}

Installation via Smithery

To install via Smithery:

smithery install --package mongo-mcp

Then add the following to your smithery.toml:

[mongo-mcp]
command = "node"
args = ["index.js"]

✨ Features

• 🔍 Check collection schemas
• 📊 Query and filter documents
• 📈 Manage indexes
• 📝 Perform document operations (insert, update, delete)

Demo Video

https://github.com/user-attachments/assets/2389bf23-a10d-49f9-bca9-2b39a1ebe654

📦 Installation

Prerequisites

• Node.js (version 18+)
• npm or yarn
• Docker (for optional sandbox environment)

Install Dependencies

Install project dependencies using the following command:

npm install

Or

yarn install

Start the Server

To start the MCP server, run:

npm start

Or

yarn start

💻 Usage Instructions

Set Up the Test Environment

Start MongoDB with Docker

You can start MongoDB in Docker using the following command:

docker run --name mcp-mongo -d mongo:latest

Then, you can connect to the MongoDB instance using the following command:

mongo --host localhost --port 27017

Create Test Data

To populate test data, run the following command:

npx mongo-mcp create-users-collection

Configure Claude Desktop

To configure Claude Desktop to use this MCP server, edit the smithery.toml file and add the following:

[mongo-mcp]
command = "node"
args = ["index.js"]

📁 Test Sandbox Data Structure

Users Collection

The users collection contains the following fields:

• _id: Document ID
• username: Username
• email: Email address
• password: Password hash
• created_at: Creation timestamp
• updated_at: Update timestamp

Example document:

{
  "_id": "63b4d207f8dfg5c1a7b1",
  "username": "exampleuser",
  "email": "user@example.com",
  "password": "$2y$10$92IXUNpkjO0rOQ5byMi.Ye4oKfW3p2.tU6VNU.0oGoeHabKN0u8F",
  "created_at": "2023-07-20T12:00:00Z",
  "updated_at": "2023-07-20T12:00:00Z"
}

Products Collection

The products collection contains the following fields:

• _id: Document ID
• name: Product name
• price: Price (in currency)
• description: Description
• category: Category
• stock: Stock quantity
• created_at: Creation timestamp

Example document:

{
  "_id": "63b4d207f8dfg5c1a7b2",
  "name": "Example Product",
  "price": 9.99,
  "description": "This is an example product.",
  "category": "electronics",
  "stock": 10,
  "created_at": "2023-07-20T12:00:00Z"
}

Orders Collection

The orders collection contains the following fields:

• _id: Document ID
• user_id: User ID (references the _id in the users collection)
• products: List of products in the order (array containing product IDs and quantities)
• total_price: Total price (in currency)
• order_date: Order date timestamp
• status: Order status (e.g., "pending", "processing", "shipped", "delivered")

Example document:

{
  "_id": "63b4d207f8dfg5c1a7b3",
  "user_id": "63b4d207f8dfg5c1a7b1",
  "products": [
    {
      "product_id": "63b4d207f8dfg5c1a7b2",
      "quantity": 2
    }
  ],
  "total_price": 19.98,
  "order_date": "2023-07-20T12:00:00Z",
  "status": "pending"
}

🛠️ Available Tools

Command List

The following are some available commands and their functions:

Index Management

• create-collection: Create a new MongoDB collection.
• drop-collection: Delete a specified MongoDB collection.
• add-index: Add an index to a specified collection.
• remove-index: Remove an index from a specified collection.

Data Operations

• insert-doc: Insert a single document into a specified collection.
• batch-insert: Batch insert multiple documents into a specified collection.
• find-docs: Query all documents in a specified collection (optional conditions).
• update-doc: Update one or more documents in a specified collection.
• delete-doc: Delete one or more documents in a specified collection.

📄 License

This project is licensed under the MIT License.

For further information, please refer to the project's full documentation.