PostgreSQL Tuning MCP server

This is a helper for people who look after a PostgreSQL database and want a hand finding what is slow. It connects your AI assistant to your database so it can read performance information and explain it in plain words. It is handy if you run reports, dashboards, or apps on PostgreSQL and queries sometimes crawl.

What this MCP server does

You ask your AI something like which queries are slowest or why a report takes so long. The AI sends that request to this helper program. The helper connects to your PostgreSQL database and reads the performance statistics that PostgreSQL already keeps. It then hands the numbers back to the AI, which explains them to you and suggests what to change, such as adding an index or adjusting a setting.

Click to zoom

What you can do with it

• List the slowest queries in your database
• Explain why a specific query is slow
• Suggest indexes that could speed up your workload
• Test a suggested index without actually creating it
• Find unused or duplicate indexes you could drop
• Check overall database health and cache hit ratios
• Watch long-running vacuum operations

Try asking your AI

• “What are the slowest queries in my database this week?”
• “Why is this query slow: SELECT * FROM orders WHERE customer_id = 42”
• “Which indexes are unused and could be removed?”
• “Would adding an index on orders(customer_id) help?”

What it gives back to you

You get back plain answers in the chat: lists of slow queries with their times, explanations of a query plan in everyday words, and suggestions for indexes or settings. It can also show health scores, table sizes, and which indexes are unused. Nothing in your data is changed unless you decide to act on the advice yourself.

Before you start

Who it's for

Database administrators, developers, and anyone who looks after a PostgreSQL database and wants help finding slow queries.