MongoDB MCP server

The MongoDB MCP Server is a bridge between Large Language Models (LLMs) and MongoDB databases, allowing you to perform database operations using natural language through the Model Context Protocol. It enables LLMs to connect to MongoDB and execute various database operations without writing complex queries manually.

Installation

Prerequisites

• Go programming environment
• MongoDB instance
• Access credentials for your MongoDB database

Setup

1. Clone the repository:

git clone https://github.com/mark3labs/mongodb-mcp-server
cd mongodb-mcp-server

2. Create a configuration file by copying the example:

cp config.yml.example config.yml

3. Modify the configuration file to match your MongoDB setup:

mongo:
  host: mongodb
  port: 27017
  user: admin
  password: 123456
  database: db

mcp:
  name: mongo-mcp-server
  version: 1.0.0
  base_url: localhost:8081
  address: ":8081"
  sse: true

Configuration Options

MongoDB Settings

• host: Your MongoDB server hostname or IP address
• port: MongoDB server port (typically 27017)
• user: MongoDB username with appropriate permissions
• password: Password for MongoDB authentication
• database: Target database name

MCP Server Settings

• name: Name of your MCP server instance
• version: Version identifier
• base_url: Base URL for server access
• address: Server listening address and port
• sse: Enable/disable Server-Sent Events (default: true)

Starting the Server

Run the server using the following command:

go run main.go

The server will start and listen on the address specified in your configuration file.

Available Tools

Query Operations

• find: Query documents with filtering and projection
• count: Count documents in a collection
• listCollections: List all available collections
• insertOne: Insert a single document into a collection
• updateOne: Update a single document
• deleteOne: Delete a single document

Index Management

• createIndex: Create a new index on a collection
• dropIndex: Remove an existing index
• indexes: List all indexes for a collection

Using with LLMs

The MongoDB MCP Server can be integrated with LLM interfaces like DeepChat. When properly configured, you can ask your LLM to perform MongoDB operations using natural language, and the MCP server will translate these requests into proper MongoDB commands.

For example, you can ask the LLM to:

• Find all documents in a specific collection
• Count documents matching certain criteria
• Create new indexes to improve query performance
• Insert, update, or delete documents

The server handles the communication between the LLM and your MongoDB database, translating natural language requests into proper database operations.

References

For more information, check the following resources:

• MongoDB Official Documentation
• mcp-go Library

How to add this MCP server to Cursor

There are two ways to add an MCP server to Cursor. The most common way is to add the server globally in the ~/.cursor/mcp.json file so that it is available in all of your projects.

If you only need the server in a single project, you can add it to the project instead by creating or adding it to the .cursor/mcp.json file.

Adding an MCP server to Cursor globally

To add a global MCP server go to Cursor Settings > MCP and click "Add new global MCP server".

When you click that button the ~/.cursor/mcp.json file will be opened and you can add your server like this:

{
    "mcpServers": {
        "cursor-rules-mcp": {
            "command": "npx",
            "args": [
                "-y",
                "cursor-rules-mcp"
            ]
        }
    }
}

Adding an MCP server to a project

To add an MCP server to a project you can create a new .cursor/mcp.json file or add it to the existing one. This will look exactly the same as the global MCP server example above.

How to use the MCP server

Once the server is installed, you might need to head back to Settings > MCP and click the refresh button.

The Cursor agent will then be able to see the available tools the added MCP server has available and will call them when it needs to.

You can also explictly ask the agent to use the tool by mentioning the tool name and describing what the function does.