How to add Postgres MCP Pro to Cursor

Read/write Postgres access plus index tuning, EXPLAIN plans, and database health analysis for AI agents. Paste the config into ~/.cursor/mcp.json and restart Cursor.

Last updated June 15, 2026 · 2.9k★ · stdio · apikey

Cursor config for Postgres MCP Pro

uvx postgres-mcp --access-mode=restricted

{
  "mcpServers": {
    "postgres-mcp-pro": {
      "command": "docker",
      "args": [
        "run",
        "-i",
        "--rm",
        "-e",
        "DATABASE_URI",
        "crystaldba/postgres-mcp",
        "--access-mode=restricted"
      ],
      "env": {
        "DATABASE_URI": "postgresql://<user>:<password>@<host>:5432/<dbname>"
      }
    }
  }
}

Requires Docker to be installed and running.

Setup steps

1Open Cursor → Settings → MCP → Add new MCP server (or edit ~/.cursor/mcp.json directly).
2Paste the Postgres MCP Pro config below into the "mcpServers" object.
3Fill in placeholder secrets, then save.
4Cursor reloads MCP servers automatically — check Settings → MCP for a green status dot.
5Ask Cursor to use one of Postgres MCP Pro's tools to confirm it's connected.

Before you start

A reachable PostgreSQL database and a connection string in the form postgresql://user:password@host:5432/dbname
Python 3.12+ with the uv or pipx package manager, OR Docker if you prefer the container image
Recommended Postgres extensions for full functionality: pg_stat_statements (query stats) and hypopg (hypothetical indexes for tuning)
An MCP-capable client (Claude Desktop, Cursor, Windsurf, etc.)

What Postgres MCP Pro can do in Cursor

list_schemas

List all schemas in the database

list_objects

List tables, views, sequences, and extensions within a schema

get_object_details

Get columns, constraints, and indexes for a table/view/object

execute_sql

Run SQL; read-only when the server is in restricted mode

explain_query

Return the execution plan, optionally simulating hypothetical indexes

get_top_queries

Report the slowest/most resource-intensive queries via pg_stat_statements

analyze_workload_indexes

Recommend indexes for the overall workload using greedy search + hypopg

analyze_query_indexes

Recommend indexes for a specific set of SQL queries (up to 10)

Security

The DATABASE_URI contains full Postgres credentials, so store it securely and prefer a least-privilege role. Use --access-mode=restricted (read-only) unless the agent genuinely needs write/DDL access.

Postgres MCP Pro + Cursor FAQ

Where is the Cursor config file?

Cursor reads MCP servers from ~/.cursor/mcp.json. Paste the Postgres MCP Pro config there under the "mcpServers" key and restart the client.

Is Postgres MCP Pro safe to use with Cursor?

Is there an npm package?

No. Postgres MCP Pro is Python-only, distributed on PyPI (pipx/uv) and as a Docker image. There is no Node/npm package.

How do I make it safe for production?

Run with --access-mode restricted, which limits operations to read-only transactions and applies resource/time limits, and connect with a least-privilege database role.

Why do I need pg_stat_statements and hypopg?

pg_stat_statements powers slow-query analysis, and hypopg lets the server simulate indexes to test their impact without creating them. Without them, tuning and top-query features are limited.

View repo Full Postgres MCP Pro page