Mysql MCP Server

🚀 MySQL MCP Server

The MySQL MCP Server provides a Model Context Protocol (MCP) interface for MySQL databases, enabling AI models to interact with MySQL databases through a structured interface.

🚀 Quick Start

The MySQL MCP Server offers an MCP interface for MySQL databases, allowing AI models to interact with them via a structured interface. It adheres to the Model Context Protocol specification and supports both Web server and CLI operating modes.

✨ Features

• MySQL database connection management
• SQL query execution
• Table list and structure inspection
• Database list and selection
• Real - time status monitoring via SSE (Server - Sent Events)
• Web interface provided for testing MCP tools
• Support for both stdio and SSE transmission methods
• Ready for Docker deployment

📦 Installation

# Global installation
npm install -g mysql-mcp

# Local installation
npm install mysql-mcp

💻 Usage Examples

Using the Server Published on Smithery.ai

The MySQL MCP Server is published on Smithery.ai and can be easily used with various AI assistants:

1. Access the server: Visit https://smithery.ai/server/@sussa3007/mysql-mcp.
2. Configure the server:
   • Set the MySQL database connection parameters:
     • MYSQL_HOST
     • MYSQL_PORT
     • MYSQL_USER
     • MYSQL_PASSWORD
     • MYSQL_DATABASE
     • MYSQL_READONLY (optional, default is True for read - only access).
3. Connect to supported AI platforms:
   • Anthropic Claude
   • Cursor AI
   • Windsurf
   • Cline
   • Witsy
   • Enconvo
   • Goose
4. Authentication: Log in to Smithery.ai to save the configuration and generate an authentication token.
5. Use in AI prompts: After connecting, you can use the MySQL tools by requesting database operations in AI conversations.

Using the Locally Installed Version

To use the locally installed version:

npm install mysql-mcp

Run the server:

mysql-mcp

By default, the server will run on http://localhost:3000. You can access /api-docs to view the API documentation.

📚 Documentation

MCP Service

API Endpoints

/ping

• Method: GET
• Description: Check if the service is available
• Response:
  • status: "ok"

/query

• Method: POST
• Path: /query
• Request Body:

  {
    "sql": "string" // SQL query to be executed
  }
  

• Response Body:

  {
    "results": [...] // Query results
  }
  

📄 License

MIT