---
title: "Build with AI agents"
description: "Point your AI coding agent at Cardzone's llms.txt, Markdown pages, or OpenAPI spec, and start from a ready prompt."
---

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

# Build with AI agents

This page lists the files an AI coding agent can read, and gives you prompts to start with.

These docs are written for coding agents as well as people. Everything below is generated from the same pages and the same OpenAPI spec that you read here. (The OpenAPI spec is a machine-readable description of every endpoint.) So agents always see the same API contract as you.

## What agents can read

| Resource | Path | What it is for |
| --- | --- | --- |
| Site index | `/llms.txt` | A list of links to every page and API operation, in the [llmstxt.org](https://llmstxt.org) format. |
| Full text | `/llms-full.txt` | All pages in one Markdown file, so an agent can load everything at once. |
| Any page as Markdown | add `index.md` to the end of a page URL, like `/guides/webhooks/index.md` | One page, without the site menus. |
| API reference index | `/reference/llms.txt` | Only the operations, webhooks, and schemas. |
| OpenAPI 3.1 spec | `/reference/openapi.json` | Typed clients and request validation. |
| Live OpenAPI spec | `https://cardzone.ai/api/b2b/openapi.json` | The same spec, served by the API itself, for tools that should always use the latest version. |

Each page also has a **Copy page** button in its header. It copies the page's Markdown, so you can paste it into a chat.

## Ready prompts

> **Tip**
>
> Test code written by an agent in the sandbox first. The [sandbox scenarios](https://docs.cardzone.ai/resources/sandbox/) cover every status and action, and they spend no money.

Source: https://docs.cardzone.ai/resources/for-ai-agents/index.mdx