PostgreSQL MCP Server

Official

Query your PostgreSQL database with natural language

postgresqlsqldatabaseofficialread-onlylocal

The official PostgreSQL MCP server from Anthropic lets your AI assistant read and query any PostgreSQL database. Ask questions in plain English — the AI translates them into SQL and returns results instantly. The server runs locally, so your data and connection credentials never leave your machine. Supports schema inspection, table listing, and arbitrary read queries. Write queries are not supported by default, keeping your data safe.

Install Config

Config file: ~/Library/Application Support/Claude/claude_desktop_config.json· Windows: %APPDATA%\Claude\claude_desktop_config.json

{
  "mcpServers": {
    "postgres": {
      "command": "npx",
      "args": [
        "-y",
        "@modelcontextprotocol/server-postgres"
      ],
      "env": {
        "POSTGRES_URL": "postgresql://user:password@localhost:5432/mydb"
      }
    }
  }
}

Required environment variables:

POSTGRES_URL — PostgreSQL connection string(e.g. postgresql://user:password@localhost:5432/mydb)

Available Tools (3)

Tool name	Description
query	Execute a read-only SQL query against the database
list_tables	List all tables in the connected database
describe_table	Get the full schema for a specific table

Example Prompts

“What tables are in my database?”
“Show me the 10 most recent orders”
“How many users signed up this week?”
“What is the schema for the products table?”
“Find all orders over $500 placed in the last 30 days”

Pros & Cons

Pros

Official Anthropic-maintained server
Read-only by default — safe for production
All data stays on your machine
Works with all major MCP clients

Cons

PostgreSQL only — not MySQL or MongoDB
No write support by default
Requires a running PostgreSQL instance

Who Is This For?

·Developers who want conversational access to their app database
·Data analysts wanting AI-assisted SQL generation
·Teams demoing data capabilities without writing queries

When to use: Use this server when you have a PostgreSQL database and want to explore your data, generate reports, or answer questions without writing SQL yourself.