PostgreSQL

Provides read-only access to PostgreSQL data via MCP for querying, listing schemas, and inspecting tables.

python

GitHub Stars

python

Language

4 months ago

First Indexed

2 months ago

Catalog Refreshed

Documentation & install

Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.

Installation

Add the following to your MCP client configuration file.

Configuration

View docs

{
  "mcpServers": {
    "psreekrishnan-postgresql-mcp": {
      "command": "C:\\\\Users\\\\sk\\\\AppData\\\\Local\\\\Programs\\\\Python\\\\Python313\\\\python.exe",
      "args": [
        "C:\\\\Workspace\\\\GitLab\\\\capacity-planning2\\\\postgresql-mcp\\\\server.py",
        "--host",
        "host.example.com",
        "--port",
        "61025",
        "--database",
        "example",
        "--username",
        "example_user",
        "--password",
        "example_pwd"
      ],
      "env": {
        "PYTHONUNBUFFERED": "1"
      }
    }
  }
}

You can connect to a PostgreSQL database through an MCP server that exposes a read-only, query-enabled interface to an Amazon Q client. This MCP server runs locally or remotely, executes approved SELECT queries, and returns results in a controlled, secure way. It is designed to be easy to install, test, and configure so you can start querying data immediately from your MCP-enabled workflows.

How to use

You interact with the PostgreSQL MCP Server through an MCP client. First you install the server dependencies, then register a local or remote MCP entry that points to the server, and finally you run queries from your MCP client. The server is configured to allow read-only access with automatic limits and a connection timeout to protect your data.

How to install

pip install -r requirements.txt

python server.py --host host.example.com --port 61025 --database example --username example_user --password example_pwd --test

# Configure Amazon Q MCP (Windows path example)
C:\Users\%USERNAME%\.aws\amazonq\mcp.json

{
  "mcpServers": {
    "local-postgres-mcp": {
      "command": "C:\\Users\\sk\\AppData\\Local\\Programs\\Python\\Python313\\python.exe",
      "args": [
        "C:\\Workspace\\GitLab\\capacity-planning2\\postgresql-mcp\\server.py",
        "--host", "host.example.com",
        "--port", "61025",
        "--database", "example",
        "--username", "example_user",
        "--password", "example_pwd"
      ],
      "env": {"PYTHONUNBUFFERED": "1"},
      "timeout": 60000
    }
  }
}

# Restart your IDE to load the new MCP server configuration

Security and usage notes

The server enforces read-only access, allowing only SELECT queries. It automatically limits result sets (e.g., 100 rows max) to prevent large data transfers, and it enforces a connection timeout to protect resources.

Available tools

query_database

Execute SELECT queries with an automatic LIMIT to protect data exposure.

describe_table

Fetch detailed information about table columns and data types.

list_tables

List tables within a specific schema or across all schemas.

Built by

VeilStrat

AI signals for GTM teams