> ## Documentation Index
> Fetch the complete documentation index at: https://docs.mightynetworks.com/llms.txt
> Use this file to discover all available pages before exploring further.

# Use the Docs with AI Tools

> Connect the developer docs to Claude, Cursor, ChatGPT, and other AI tools through the docs MCP server and machine-readable formats

If an AI assistant helps you build your integration, point it at this site directly. Everything here — the guides, the full [Headless API](/headless-api) GraphQL reference, and the [Admin API](/admin-api) reference — is published in formats built for AI tools, so your assistant works from the current docs instead of stale training data or scraped HTML.

## Docs MCP server

The documentation is available as a [Model Context Protocol](https://modelcontextprotocol.io/) (MCP) server:

```
https://docs.mightynetworks.com/mcp
```

Once connected, your AI tool can search the developer docs, retrieve any page as clean Markdown, run read-only queries across the documentation corpus, and report documentation problems straight to the docs team. The server requires no authentication and is read-only — it exposes only the published content of this site.

<Note>
  This server answers questions about the **documentation**. It is separate from any MCP connection your own Mighty Network offers for its content — that one is managed in Network Admin and authenticates as a member of the Network.
</Note>

### Connect your tool

<Tabs>
  <Tab title="Claude Code">
    Add the server from your terminal:

    ```bash theme={null}
    claude mcp add --transport http mighty-networks-docs https://docs.mightynetworks.com/mcp
    ```
  </Tab>

  <Tab title="Claude">
    In the Claude web or desktop app, go to **Settings** > **Connectors** > **Add custom connector**, name it, and paste the server URL. See the [Claude custom connector guide](https://support.claude.com/en/articles/11175166-get-started-with-custom-connectors-using-remote-mcp) for a walkthrough.
  </Tab>

  <Tab title="Cursor">
    Add the server to `.cursor/mcp.json` in your project (or `~/.cursor/mcp.json` globally):

    ```json theme={null}
    {
      "mcpServers": {
        "mighty-networks-docs": {
          "url": "https://docs.mightynetworks.com/mcp"
        }
      }
    }
    ```
  </Tab>

  <Tab title="VS Code">
    Add the server to `.vscode/mcp.json` in your workspace:

    ```json theme={null}
    {
      "servers": {
        "mighty-networks-docs": {
          "type": "http",
          "url": "https://docs.mightynetworks.com/mcp"
        }
      }
    }
    ```
  </Tab>

  <Tab title="ChatGPT">
    Enable **Developer Mode** under **Settings** > **Apps & Connectors**, then create a connector with the server URL. See the [ChatGPT connectors guide](https://help.openai.com/en/articles/12584461-developer-mode-and-mcp-apps-in-chatgpt).
  </Tab>
</Tabs>

Any other MCP-capable client works the same way: it's a streamable HTTP server at the URL above, with no authentication step.

## Every page as Markdown

Append `.md` to any page's URL to get the raw Markdown instead of HTML:

```bash theme={null}
curl https://docs.mightynetworks.com/headless-api.md
```

You can also use the contextual menu in the top-right corner of every page to copy the page as Markdown, open it directly in Claude, ChatGPT, or Perplexity, or copy the MCP server address.

## llms.txt

The site publishes the [llms.txt](https://llmstxt.org/) convention for AI crawlers and one-shot ingestion:

* [`/llms.txt`](https://docs.mightynetworks.com/llms.txt) — an index of every page with one-line descriptions, so a tool can find and fetch exactly the pages it needs
* [`/llms-full.txt`](https://docs.mightynetworks.com/llms-full.txt) — the entire documentation in a single Markdown file

## Generating typed clients

For working with the Headless API schema itself — rather than the docs about it — fetch the SDL from the unauthenticated schema endpoint and feed it to your code generator. See [Fetching the schema](/headless-api#fetching-the-schema).