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

# Discovery files

> The public files that tell AI apps and crawlers where to find the Private Clubs MCP server and API

Private Clubs publishes a few files at fixed addresses on [https://www.theprivateclubs.com](https://www.theprivateclubs.com). AI apps, agents and crawlers read them to find the Private Clubs MCP server and API without help.

## Files

| File | What it holds |
| - | - |
| [`/mcp.json`](https://www.theprivateclubs.com/mcp.json) | The MCP server card: the name Private Clubs, version, logo, endpoint, transport, the five tools, the server key `private-clubs`, `auth` set to `none`, and the setup page. |
| [`/openapi.json`](https://www.theprivateclubs.com/openapi.json) | The REST API spec. See [API overview](/api/openapi). |
| [`/llms.txt`](https://www.theprivateclubs.com/llms.txt) | A site overview for AI tools, regenerated hourly. It notes that the API and the MCP tools take a full state name or a two-letter code, such as `Florida` or `FL`. |
| [`/ai/context.md`](https://www.theprivateclubs.com/ai/context.md) | A short Markdown summary of Private Clubs for AI tools: club counts, how fees are shown, waitlist shares and key URLs, with `/api/v1/clubs` as the club data API. It carries no fee figures. |
| [`/mcp/setup.md`](https://www.theprivateclubs.com/mcp/setup.md) | Setup steps an agent reads and follows for its own app. |
| [`/robots.txt`](https://www.theprivateclubs.com/robots.txt) | Crawler rules, open to AI crawlers. |
| [`/data/license`](https://www.theprivateclubs.com/data/license) | The data license, a web page: non-commercial use with attribution to Private Clubs, AI training with citation, the attribution formats, and [support@theprivateclubs.com](mailto:support@theprivateclubs.com) for commercial use. |

`/mcp.json`, `/llms.txt`, `/ai/context.md` and `/mcp/setup.md` are served with `Cache-Control: public, max-age=3600`, so a cached copy can be up to an hour old.

## For AI crawlers

`robots.txt` opens the site to search, AI input and AI training. Its catch-all group carries this line:

```text theme={"system"}
Content-Signal: search=yes, ai-input=yes, ai-train=yes
```

That group allows `/` and disallows a short list of private paths, including sign-in, account, dashboard, admin and `/api/`. Named AI crawlers each get their own `Allow: /` group. They include OAI-SearchBot, ChatGPT-User, GPTBot, Claude-SearchBot, Claude-User, ClaudeBot, PerplexityBot, Perplexity-User, Google-Extended, Meta-WebIndexer, Amazonbot and Applebot-Extended.

## Which file to use

* **Connect an AI app:** point it at `/mcp/setup.md`, or follow [Install Private Clubs](/mcp/connect).
* **Find the MCP server from code:** read `/mcp.json`.
* **Build on the REST API:** read `/openapi.json`.
* **Give an AI tool a site overview:** use `/llms.txt`, or `/ai/context.md` for a shorter summary.
* **Check crawler rules:** read `/robots.txt`.
* **Check the terms of use:** read `/data/license`.

Next: [Conventions](/reference/conventions) and [API overview](/api/openapi).


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.