MySQL MCP Server

English | 中文

Overview

A Model Context Protocol (MCP) server implementation that provides database interaction capabilities through MySQL. This server enables running SQL queries, creating tables, and exploring database schema information.

Components

Tools

The server offers five core tools:

Query Tools

• read_query

  • Execute SELECT queries to read data from the database

  • Input:
    * query (string): The SELECT SQL query to execute

  • Returns: Query results as array of objects

• write_query

  • Execute INSERT, UPDATE, or DELETE queries

  • Input:
    * query (string): The SQL modification query

  • Returns: { affected_rows: number }

• create_table

  • Create new tables in the database

  • Input:
    * query (string): CREATE TABLE SQL statement

  • Returns: Confirmation of table creation

Schema Tools

• list_tables

  • Get a list of all tables in the database

  • No input required

  • Returns: Array of table names

• describe_table

  • View schema information for a specific table

  • Input:
    * table_name (string): Name of table to describe

  • Returns: Array of column definitions with names and types

Installation

Prerequisites

• Python 3.10+

• MySQL database

• Required Python packages:

  • mcp (Model Context Protocol)

  • sqlalchemy

  • pymysql (or another MySQL driver)

  • python-dotenv

  • uvicorn (for HTTP transport)

Setup with Conda

First, create and activate a conda environment:

Then, install the required dependencies:

Configuration

You can configure the server using:

Environment Variables File (.env)

1. Copy the .env.template file and rename it to .env:

2. Edit the .env file with your configuration:

Command Line Arguments

You can also override configuration with command line arguments:

Usage

Starting the Server

Testing with MCP Inspector

You can test the server using the MCP Inspector tool:

This will start the server and allow you to interactively test the available tools.

Example Workflow

1. Start the server with your MySQL database connection

2. Connect an AI model to the server using the MCP client

3. Use the list_tables tool to see available tables

4. Create tables with create_table if needed

5. Insert data with write_query

6. Query data with read_query

Usage with Claude Desktop

uv

Add the server to your claude_desktop_config.json:

Docker

Add the server to your claude_desktop_config.json:

Package Installation

You can also install the package using pip:

In Cursor IDE

Cursor is an AI-assisted IDE. You can integrate this MCP server with Cursor to query MySQL databases directly during coding.

Setup in Cursor

1. Start the MCP server

   python src/mysql/server.py

2. Configure MCP in Cursor settings

   Add your MCP server URL:

   http://localhost:8000

3. Use Cursor commands to access MCP

   In the Cursor editor, use:

   /mcp mysql-query {"query": "SELECT * FROM users LIMIT 5"}

   For parameterized queries:

   /mcp mysql-query {"query": "SELECT * FROM users WHERE age > :min_age", "params": {"min_age": 30}}

API Reference

Input Format

Output Format

Security Considerations

• This server should be run in a trusted environment as it allows arbitrary SQL queries

• In production, implement proper access control and input validation

• Consider limiting the types of SQL commands that can be executed

• Important: Do not commit .env files containing sensitive information to version control

Development

Project Structure

• src/mysql/server.py: Main server implementation

• pyproject.toml: Package configuration

• README.md: This documentation

Adding New Features

To extend the server with new capabilities:

1. Add new tools using the @mcp.tool() decorator

2. Implement the tool logic using the MySQLDatabase class

3. Update the documentation to reflect the new capabilities

License

This MCP server is licensed under the MIT License. This means you are free to use, modify, and distribute the software, subject to the terms and conditions of the MIT License.