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

> ## Agent Instructions
> State which API version you are using. V1 uses https://esa.acreblitz.com/api/v1; V2 uses https://esa-v2.acreblitz.com/api/v2.
> Read /llms.txt for links to both versions. For V2 bulk integration, read bulk submission, status, results, and /v2/errors before constructing requests.

# Read docs with an AI agent

> Fetch documentation directly by URL, with indexes and complete Markdown for V1 and V2

Give your agent a documentation URL and ask it to fetch the page. You do not need to copy the page into the chat. Reading public documentation requires no API key. Calling the ESA API still requires your provider key.

## Discover pages

Start at [llms.txt](https://docs.acreblitz.com/llms.txt). It links to every documented V1 and V2 page, including every endpoint. Choose the version before building requests:

| Version | API host and prefix | Documentation |
| - | - | - |
| V1 | `https://esa.acreblitz.com/api/v1` | Existing pages outside `/v2/` |
| V2 | `https://esa-v2.acreblitz.com/api/v2` | Pages under `/v2/` |

```bash theme={null}
curl --fail --location https://docs.acreblitz.com/llms.txt
```

## Read a page directly

Append `.md` to a page URL to fetch its Markdown, including parameters and examples:

```bash theme={null}
curl --fail --location \
  https://docs.acreblitz.com/v2/api-reference/endpoint/bulk-submit.md
```

The hosted site also supports requesting Markdown with an `Accept` header:

```bash theme={null}
curl --fail --location --header 'Accept: text/markdown' \
  https://docs.acreblitz.com/v2/api-reference/endpoint/bulk-submit
```

For full context, follow the **All V1 and V2 documentation** link in [llms.txt](https://docs.acreblitz.com/llms.txt). That link includes a version parameter for `llms-full.txt` so you receive the current release even when a CDN retains an older copy of the unversioned URL. It contains both versions and expands shared input documentation so agents can read required fields without following source-code imports. Prefer individual pages when the agent has a limited context window.

## Example agent task

> Fetch [https://docs.acreblitz.com/llms.txt](https://docs.acreblitz.com/llms.txt) and use only the V2 documentation. Read the bulk submission, job status, results, and error pages directly. Build a request with shared products and crops, then override products and crops for one field. Explain the effective values for each field and handle both HTTP errors and failed items. Read the URLs yourself; do not ask me to paste their contents.

An agent needs a web-fetch or browsing tool to use these links. If its tool cannot fetch a page, it should report that limitation and try the Markdown URL or complete-content file. It should not guess endpoint behavior from an unread page.