Docker provides two ways to use the Geonode MCP Server.

Docker Desktop provides a graphical interface for managing MCP servers and AI clients.
Docker CLI runs a lightweight MCP bridge directly from the terminal.

Choose the option that best fits your workflow.

Before you begin

Make sure you've completed the Before You Start guide and have:

A Geonode API key
The Geonode MCP endpoint
Docker installed

Option 1 — Docker Desktop

Docker Desktop includes the MCP Toolkit, which allows you to manage MCP servers and connect them to supported AI clients.

Step 1 — Install Docker Desktop

Download and install Docker Desktop for your operating system.

Launch Docker Desktop and ensure the Docker Engine is running.

Step 2 — Open the MCP Toolkit

From the Docker Desktop sidebar, open:

Models → MCP Toolkit

This is where Docker manages MCP servers and connected AI clients.

Docker Desktop MCP Toolkit

Step 3 — Connect Your AI Client

Open the Clients tab.

Choose the AI client you want to use.

For example:

Cursor
Continue.dev
Codex
Gemini CLI
Goose

Click Connect next to your client.

Docker configures the client to communicate with the MCP Toolkit.

Connect Your AI Client

Step 4 — Restart Your AI Client

After connecting the client, restart it.

Once restarted, Docker automatically exposes the configured MCP servers to the client.

Step 5 — Verify the Connection

Open your AI client's MCP settings.

You should see:

a green status indicator
the configured MCP server
the available Geonode tools

Verify the Connection

Option 2 — Docker CLI

If you prefer using the terminal, Docker can run a lightweight MCP bridge without any additional installation.

Step 1 — Verify Docker

Open a terminal.

Run:

docker --version

If Docker is installed correctly, the installed version is displayed.

Step 2 — Start the MCP Bridge

Run:

docker run -i --rm \
  -e GEONODE_API_KEY=YOUR_API_KEY \
  node:20-alpine \
  npx -y mcp-remote https://scraper.geonode.io/mcp \
  --header "X-Api-Key:${GEONODE_API_KEY}"

Replace YOUR_API_KEY with your Geonode API key.

This command:

starts a temporary Docker container
downloads mcp-remote
connects to the Geonode MCP Server
creates a local MCP bridge over STDIO

Step 3 — Verify the Bridge

When the bridge starts successfully, you'll see output similar to:

Using transport strategy: http-first
Connected to remote server using StreamableHTTPClientTransport
Local STDIO server running
Proxy established successfully
Press Ctrl+C to exit

The bridge is now ready to receive requests.

Verify the Connection

Step 4 — Configure Your AI Client

Configure your AI client to use the local Docker bridge.

Once connected, the client automatically gains access to the available Geonode MCP tools.

Start Using Geonode MCP

You can now ask your AI assistant to perform tasks such as:

Extract a web page

Extract the content from https://geonode.com

Summarize a page

Summarize https://geonode.com

Crawl a website

Crawl https://docs.geonode.com

Process multiple pages

Extract and summarize these URLs:

https://example.com/page1
https://example.com/page2
https://example.com/page3

View extraction statistics

Show my Geonode extraction statistics.

Stopping the Docker Bridge

To stop the Docker bridge, press:

Ctrl + C

Since the container is started with the --rm flag, Docker automatically removes it after it exits.

Docker MCP