MCP.so
登录
服务器

MongoDB MCP Server 🚀

@syedasad-kiwi

A powerful Model Context Protocol (MCP) server implementation for MongoDB integration

概览

What is MongoDB MCP Server?

A Model Context Protocol (MCP) server implementation for MongoDB integration, providing standardized interaction with MongoDB databases. It supports operations on databases, collections, and documents with async/await patterns and error handling, and is built for use with AI assistants and MCP-compatible IDEs.

How to use MongoDB MCP Server?

Run with uvx mongodb-mcp-bridge (Python) or npx mongodb-mcp-bridge (Node.js); add --transport=sse for remote connections. Set the required MONGODB_URI environment variable and optionally MONGODB_DB and MCP_PORT. Docker support is available via docker build and docker run. Configure in VS Code, Cursor, or Windsurf using the provided JSON snippets.

Key features of MongoDB MCP Server

  • Full MongoDB CRUD operations support
  • Secure connection handling with MongoDB Atlas
  • Async/await patterns for optimal performance
  • Comprehensive error handling
  • Docker support for easy deployment
  • Query execution with proper type hints

Use cases of MongoDB MCP Server

  • Quick connectivity test via the ping tool
  • List all databases in a MongoDB instance
  • Query documents in a collection with filters and limits
  • Integrate MongoDB operations into AI assistant workflows

FAQ from MongoDB MCP Server

What runtime or dependencies are required?

Python with uvx or Node.js with npx; no additional manual installation of drivers is needed.

How does it connect and authenticate to MongoDB?

Authentication is handled via the MONGODB_URI environment variable, which contains the MongoDB connection string including credentials.

What transport does the server use?

By default it uses standard MCP transport; the --transport=sse flag enables SSE (Server-Sent Events) transport for remote connections.

Where does my data live?

All data remains in the connected MongoDB database. The server only executes queries and returns results; it does not store data locally.

What are the available operations?

The server provides ping, list_databases, and find_documents tools for connectivity testing, database enumeration, and filtered document queries.

来自「数据库」的更多内容