Before You Start
Before configuring Geonode MCP, make sure you have everything you need. This guide covers the API key, authentication, endpoint, and the tools your AI assistant can access once connected.
Prerequisites
Before continuing, you'll need:
- A Geonode account.
- A valid Geonode API key.
- An MCP-compatible client such as Claude Desktop, Claude Code, Cursor, Windsurf, or Docker.
Get Your API Key
Every request sent through the Geonode MCP Server requires an API key for authentication.
To create or view your API key:
- Sign in to your Geonode dashboard.
- Open API Keys.
- Copy an existing key or create a new one.
- Store your API key securely.
Keep your API key private
Never share your API key or commit it to a public repository. Anyone with your API key can make requests on your behalf.
MCP Endpoint
Configure your MCP client to connect to the following endpoint:
https://scraper.geonode.io/mcpAuthentication
Geonode MCP authenticates requests using your API key.
Most MCP clients send the key through the X-Api-Key HTTP header.
Some clients cannot configure custom headers. In those cases, you can provide the key as the api_key argument when calling a tool.
Using the X-Api-Key header is the recommended approach, and all setup guides in this documentation use this method.
Replace YOUR_API_KEY
Replace YOUR_API_KEY with your own Geonode API key in every configuration example throughout this documentation.
Available MCP Tools
Once connected, your AI assistant can automatically use the following Geonode tools.
Content Extraction
| Tool | Description |
|---|---|
extract | Extract content from a single web page in Markdown or HTML. Supports JavaScript rendering, residential proxies, geo-targeting, and custom headers. |
job | Retrieve the result of a previously submitted asynchronous extraction job. |
jobs | List your extraction jobs and filter them by status, URL, date, or output format. |
Batch Processing
| Tool | Description |
|---|---|
batch | Extract up to 1,000 URLs in a single asynchronous batch job. |
batch_status | Monitor the progress of a batch job and retrieve paginated results. |
cancel_batch | Cancel a running batch job. |
Website Crawling
| Tool | Description |
|---|---|
crawl | Crawl an entire website starting from a seed URL. Supports breadth-first crawling, configurable depth, and domain filtering. |
crawl_status | Check the progress of a crawl job and retrieve discovered pages. |
cancel_crawl | Cancel a running crawl job. |
Usage Statistics
| Tool | Description |
|---|---|
statistics | Retrieve usage statistics such as extraction count, success rate, and token usage. |
How Your AI Uses These Tools
You don't need to manually choose which tool to use.
Instead, simply describe the task, and your AI assistant selects the appropriate Geonode tool automatically.
For example:
| You ask | AI uses |
|---|---|
| "Extract the content from this page." | extract |
| "Process these 500 URLs." | batch |
| "Crawl this documentation website." | crawl |
| "Check whether my extraction job has finished." | job or batch_status |
| "Show my extraction statistics." | statistics |
Next Steps
You're now ready to configure Geonode MCP.
Choose the setup guide for your preferred client:
- Claude Desktop
- Claude Code
- Cursor
- Windsurf
- Smithery
- Docker MCP