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

# Docs for LLMs

> Machine-readable versions of this documentation, including the whole site as a single file, for AI assistants and coding agents.

This documentation is published in formats that AI assistants can read
directly, so you can hand an LLM the full Pipecorn API reference instead of
pasting pages one at a time.

## The files

| File                                                       | What it contains                                                                                                                                                                 | Use it when                                                              |
| ---------------------------------------------------------- | -------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | ------------------------------------------------------------------------ |
| [`llms-full.txt`](https://docs.pipecorn.com/llms-full.txt) | **Every page of this documentation concatenated into one file**, including endpoint references, error codes, and the [credits vs limits](/api-reference/credits-vs-limits) guide | You want the model to know the whole API in one shot                     |
| [`llms.txt`](https://docs.pipecorn.com/llms.txt)           | A short index: one line per page with a link and a summary                                                                                                                       | You want a small context footprint and let the model fetch what it needs |
| [`agents.md`](https://docs.pipecorn.com/agents.md)         | Agent-facing guide: which surface to use (MCP, REST, CLI), conventions to follow, auth                                                                                           | You are building an agent that acts on Pipecorn, not just reads about it |

Both `llms.txt` files follow the [llms.txt convention](https://llmstxt.org)
and are regenerated on every docs deploy, so they always match the pages you
see here.

```text theme={null}
https://docs.pipecorn.com/llms-full.txt
https://docs.pipecorn.com/llms.txt
https://docs.pipecorn.com/agents.md
```

## How to use them

<Tabs>
  <Tab title="Claude or ChatGPT">
    Paste the `llms-full.txt` URL into the conversation and ask your question.
    Both assistants fetch the file and answer from it. For long sessions,
    attach it to a Project so it stays in context.
  </Tab>

  <Tab title="Claude Code">
    Ask Claude Code to read the file, or point it at the URL in a prompt:

    ```text theme={null}
    Read https://docs.pipecorn.com/llms-full.txt, then write a script that
    enriches the contacts in leads.csv with the Pipecorn bulk enrichment endpoint.
    ```

    Add the URL to your project's `CLAUDE.md` if you integrate Pipecorn regularly.
  </Tab>

  <Tab title="Cursor">
    Open **Settings → Features → Docs**, click **Add new doc**, and paste
    `https://docs.pipecorn.com/llms-full.txt`. It then shows up under `@Docs`
    in chat.
  </Tab>

  <Tab title="Your own agent">
    Fetch `llms-full.txt` at build time or on first use and put it in the
    system prompt or a retrieval index. Fetch `llms.txt` instead when context
    is tight, and let the agent follow links.
  </Tab>
</Tabs>

## Reading vs acting

These files let a model **read** about the API. If you want the model to
**call** Pipecorn (search companies, extract leads, enrich contacts) from a
chat, connect the [MCP server](/mcp/overview) instead. The two work well
together: the MCP server exposes the tools, and `llms-full.txt` explains the
concepts behind them, such as the difference between
[credits and limits](/api-reference/credits-vs-limits).

Machine-readable discovery for the MCP server lives at
[`/.well-known/mcp.json`](https://docs.pipecorn.com/.well-known/mcp.json).