# Docs for AI Agents

> Give AI assistants and agents the Stacksync docs as Markdown, through llms.txt, MCP or a question API

Source: https://docs.stacksync.com/docs-for-ai

The Stacksync documentation is built to be read by AI assistants and agents as well as people. Every page is available as clean Markdown, and the whole site can be searched and queried from your AI tools.

| What you need                                                                                   | Use                                        |
| ----------------------------------------------------------------------------------------------- | ------------------------------------------ |
| Let an AI assistant (Claude, Cursor, VS Code, ChatGPT…) search and read the docs while it works | [MCP server](/docs-for-ai/mcp-server)      |
| Get an answer to a question, written from the docs, from a script or agent                      | [Ask API](/docs-for-ai/ask-api)            |
| Give a model the docs as context                                                                | [`llms.txt` and `llms-full.txt`](#llmstxt) |
| Read one page as Markdown                                                                       | [Markdown pages](#markdown-pages)          |

### llms.txt

Two files follow the [llms.txt](https://llmstxt.org) standard:

* [`/llms.txt`](https://docs.stacksync.com/llms.txt): an index of every page, with its title, description and a link to its Markdown, grouped by section. Start here when an agent needs to find the right page.
* [`/llms-full.txt`](https://docs.stacksync.com/llms-full.txt): the full content of every page in one Markdown file, for loading the whole documentation into a model's context or a retrieval index.

```bash
curl https://docs.stacksync.com/llms.txt
```

### Markdown pages

Every docs page is also available as Markdown. Add `.md` to its URL:

```bash
curl https://docs.stacksync.com/two-way-sync/quick-start.md
```

The home page is at `/index.md`.

Agents can also request any page URL with an `Accept: text/markdown` header and get Markdown instead of HTML:

```bash
curl -H "Accept: text/markdown" https://docs.stacksync.com/two-way-sync/quick-start
```

Each page's Markdown starts with its title, description and source URL, followed by the content. Every HTML page also links to its Markdown version with `<link rel="alternate" type="text/markdown">`.

<Callout type="info">
  You can also copy any page as Markdown, or open it in ChatGPT or Claude, from the buttons under the page title.
</Callout>

### Crawlers

[`robots.txt`](https://docs.stacksync.com/robots.txt) allows AI crawlers to index, quote and train on the docs (`Content-Signal: ai-train=yes, search=yes, ai-input=yes`), and [`sitemap.xml`](https://docs.stacksync.com/sitemap.xml) lists every page.