Mysql MCP Server

🚀 MySQL MCP Server

An MCP server for interacting with MySQL databases.

This server supports executing read-only queries (query) and write queries with final rollback (test_execute).

🚀 Quick Start

✨ Features

• Supports read-only queries (query).
• Supports write queries with final rollback (test_execute).

📦 Installation

Environment Variables

Add the following environment variables to the ~/.mcp/.env file:

MYSQL_HOST=host.docker.internal  # hostname to access the host service
MYSQL_PORT=3306
MYSQL_USER=root
MYSQL_PASSWORD=your_password

⚠️ Important Note

host.docker.internal is a special DNS name for Docker containers to access host services. Use this setting when the MySQL server is running on your host machine. If connecting to another MySQL server, replace it with the appropriate hostname.

mcp.json Configuration

{
  "mcpServers": {
    "mysql": {
      "command": "docker",
      "args": [
        "run",
        "-i",
        "--rm",
        "--add-host=host.docker.internal:host-gateway",
        "--env-file",
        "/Users/username/.mcp/.env",
        "ghcr.io/xiangma9712/mcp/mysql"
      ]
    }
  }
}

💻 Usage Examples

Starting the Server

docker run -i --rm --add-host=host.docker.internal:host-gateway --env-file ~/.mcp/.env ghcr.io/xiangma9712/mcp/mysql

⚠️ Important Note

If you are using OrbStack, host.docker.internal is automatically supported, so you can omit the --add-host parameter. Docker Desktop usually supports this setting automatically, but for better reliability, it is recommended to add the --add-host parameter.

Available Commands

Basic Usage

1. Execute a Read-Only Query

{
  "type": "query",
  "payload": {
    "sql": "SELECT * FROM your_table"
  }
}

Response:

{
  "success": true,
  "data": [
    {
      "id": 1,
      "name": "example"
    }
  ]
}

2. Test Query Execution

{
  "type": "test_execute",
  "payload": {
    "sql": "UPDATE your_table SET name = 'updated' WHERE id = 1"
  }
}

Response:

{
  "success": true,
  "data": "The UPDATE SQL query can be executed."
}

3. List Tables

{
  "type": "list_tables"
}

Response:

{
  "success": true,
  "data": ["table1", "table2", "table3"]
}

4. Describe Table Structure

{
  "type": "describe_table",
  "payload": {
    "table": "your_table"
  }
}

Response:

{
  "success": true,
  "data": [
    {
      "Field": "id",
      "Type": "int(11)",
      "Null": "NO",
      "Key": "PRI",
      "Default": null,
      "Extra": ""
    },
    {
      "Field": "name",
    }
  ]
}