MySQL MCP Server

What is MySQL MCP Server?

MySQL MCP Server is a read-only Model Context Protocol (MCP) server for interacting with MySQL and PostgreSQL databases. It allows MCP-compatible clients to inspect schemas, retrieve metadata, and execute read-only SQL queries safely.

How to use MySQL MCP Server?

Install Node.js and npm, clone the repository, run npm install and npm run build. Configure environment variables (MYSQL_HOST, MYSQL_USER, MYSQL_PASSWORD, MYSQL_PORT, PG_HOST, PG_USER, PG_PASSWORD, PG_PORT, PG_DATABASE) in a shell script and make it executable. Integrate with MCP clients like Cursor by pointing to the script.

Key features of MySQL MCP Server

• Read-only SQL query execution.
• Schema inspection and metadata retrieval.
• Visualize schema in JSON, DOT, Mermaid.
• Compare and detect schema changes.
• Discover foreign key relationships.
• Retrieve database performance metrics.

Use cases of MySQL MCP Server

• Database schema exploration for documentation.
• Running read-only analytics queries via AI assistants.
• Comparing two database schemas for migration planning.
• Visualizing table relationships for ER diagrams.
• Monitoring global database performance metrics.

FAQ from MySQL MCP Server

What databases does it support?

MySQL and PostgreSQL.

Is it safe to use in production?

Yes, all operations are read-only to prevent data modification.

What are the prerequisites?

Node.js, npm, and access to a MySQL or PostgreSQL database.

How do I integrate it with Cursor?

Create a shell script with environment variables and configure Cursor’s mcpServers to point to that script.

Does it support query history?

No, the get_query_history tool is a placeholder and not implemented. The server does not provide query-specific history.