MySQL MCP Tools

MySQL MCP Tools

This is a Model Context Protocol (MCP) server that provides tools for querying and managing a MySQL database.

Features

The server provides the following tools:

1. query_database - Execute SELECT queries on the MySQL database
2. execute_sql - Execute INSERT, UPDATE, or DELETE operations
3. list_tables - List all tables in the database
4. describe_table - Get the structure/schema of a specific table
5. test_connection - Test the database connection
6. query_to_csv - Execute a SELECT query and save results to a CSV file
7. query_to_csv_string - Execute a SELECT query and return results as a CSV string

Resources

The server also provides these resources:

• greeting://{name} - Get a personalized greeting
• database://tables - Get all tables from the database with metadata

Database Configuration

The database connection is configured using environment variables in the .env file:

2. Configure your database credentials in the .env file
3. Run the MCP server:

Usage Examples

Query Database

Execute a SELECT query:

Execute SQL Operations

Execute INSERT, UPDATE, or DELETE:

List Tables

Get all tables in the database:

Describe Table

Get table structure:

Test Connection

Check database connectivity:

Export Query Results to CSV

Save query results to a CSV file:

Get Query Results as CSV String

Get query results as CSV string in memory:

MCP Client Configuration

To use this MCP server with MCP clients, you need to configure the client to connect to this server.

Configuration File (mcp.json)

Create or update your MCP client configuration file (usually mcp.json or similar):

Using System Python:

Using Virtual Environment Python:

Using Conda Environment:

Claude Desktop Configuration

For Claude Desktop, add this configuration to your claude_desktop_config.json:

Using System Python:

Using Virtual Environment Python:

VS Code MCP Extension

If using VS Code with an MCP extension:

1. Open VS Code settings
2. Search for "MCP" settings
3. Add the server configuration:
   • Name: database-tools
   • Command: python (or full path to your Python executable)
   • Args: ["/path/to/your/mcptools/database_mcp.py"]
   • Working Directory: /path/to/your/mcptools

Finding Your Python Environment Path

To find the correct Python path for your environment:

For Virtual Environment:

For Conda Environment:

For System Python:

Example Output:

• Virtual Environment: /home/user/myproject/venv/bin/python
• Conda Environment: /home/user/miniconda3/envs/myenv/bin/python
• System Python: /usr/bin/python3

Available Resources

The server also provides resources that can be accessed by MCP clients:

• greeting://{name} - Get a personalized greeting
• database://tables - Get all tables from the database with metadata

Troubleshooting Client Configuration

1. Path Issues: Ensure all paths in the configuration are absolute paths
2. Python Environment: Make sure the Python environment has all required dependencies installed
3. Permissions: Ensure the MCP client has permission to execute the server script
4. Database Connection: Verify the .env file is properly configured with database credentials
5. Server Status: Check if the server starts without errors by running python database_mcp.py manually

Security Notes

• The database credentials are stored in environment variables for better security
• Use parameterized queries to prevent SQL injection
• Ensure your MySQL server is properly secured

Dependencies

• mcp - Model Context Protocol framework
• mysql-connector-python - MySQL database connector
• python-dotenv - Environment variable management