Supabase Self Host

🚀 Query Supabase Using Your Own MCP on Docker Host on macOS

This guide explains how to set up a local MCP (Model Context Protocol) server instance on macOS to query your Supabase PostgreSQL database.

🚀 Quick Start

Prerequisites:

• Have a Supabase account and project credentials (database password, project ID, and connection information).
• Have Docker Desktop for macOS installed (https://www.docker.com/products/docker-desktop/).

Step 1: Install Docker

Download and install Docker Desktop for macOS following the instructions on the Docker official website.

Step 2: Create a docker-compose.yml File

1. Open your preferred text editor.
2. Create a new file named "docker-compose.yml".
3. Copy and paste the following content into the file:

version: '3.8'
services:
  db:
    image: supabase/postgres:15.8.1.049
    container_name: supabase-db
    restart: unless-stopped
    environment:
      POSTGRES_PASSWORD: Your PostgreSQL password
      POSTGRES_DB: Your database name
    ports:
      - "5432:5432"
    volumes:
      - db-data:/var/lib/postgresql/data

  mcp:
    image: node:18
    container_name: mcp-supabase
    working_dir: /app
    # This command is used to run the MCP server.
    command: >
      sh -c "npx -y @modelcontextprotocol/server-postgres 'postgres://postgres:Your PostgreSQL password@db:5432/Your database name'"
    depends_on:
      - db
    ports:
      - "3000:3000"
    environment:
      NODE_ENV: production

volumes:
  db-data:

Note:

• Replace "Your PostgreSQL password" with your actual password.
• Replace "Your database name" with your actual database name.

This file defines two services:

• db: Runs the official Supabase PostgreSQL container. It exposes port 5432 and uses a Docker volume named "db-data" to persist data.
• mcp: Runs the MCP server via npx using a Node 18 container. It connects to the PostgreSQL container and exposes on port 3000.

Step 3: Configure Your Environment

Ensure that the Supabase connection information (such as database password and name) in the docker-compose.yml file has been correctly updated. You can also use a separate .env file to manage these variables.

Step 4: Start the Docker Containers

1. Open the terminal.
2. Navigate to the directory containing the docker-compose.yml file.
3. Run the following command to start the containers in detached mode:

docker-compose up -d

This command will pull the required images (if not already downloaded), create the containers, and run them as background tasks.

Step 5: Verify the Services

Execute the following command to check if all services are running properly:

docker-compose ps

You should see the "supabase-db" and "mcp-supabase" containers with a status of "Up". If you need to troubleshoot, view the logs:

docker-compose logs -f

Step 6: Interact with the MCP Server

After the MCP server is running, you can access it via a browser or an API client:

http://localhost:3000

Interact with the Supabase database using natural language queries from here. This endpoint can also be integrated with other compatible AI tools to perform SQL operations.

Conclusion

You have successfully set up a self-hosted MCP server on macOS to query your Supabase PostgreSQL database. This setup can be extended or integrated with other services in your development environment to enable natural language queries and build custom integrations.

Happy coding!