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

# LLM & AI Access

> How AI agents and coding assistants can discover and use the urltodata API.

urltodata is designed for consumption by both humans and AI agents. Two machine-readable interfaces are available alongside this documentation.

## llms.txt

A plain-text Markdown file at `/llms.txt` following the [llmstxt.org](https://llmstxt.org) specification. This gives LLMs a concise overview of every endpoint, its parameters, and expected responses — optimized for minimal token usage.

```bash theme={null}
curl https://api.urltodata.ai/llms.txt
```

Use this when you want an AI coding assistant (Cursor, Copilot, Claude Code) to understand the API quickly. Paste the content into your prompt or reference it as context.

## OpenAPI spec

A full OpenAPI 3.1 specification at `/openapi.json` with schemas, parameter definitions, and response types:

```bash theme={null}
curl https://api.urltodata.ai/openapi.json
```

AI agent frameworks that support OpenAPI function calling can consume this directly to generate typed API clients.

## Choosing the right interface

| Use case                              | Interface               |
| ------------------------------------- | ----------------------- |
| AI coding assistant needs API context | `llms.txt`              |
| Generate a typed client or SDK        | `openapi.json`          |
| Human developer exploring endpoints   | This documentation site |

## Token efficiency

The `llms.txt` file is optimized for LLM context windows — \~5KB of clean Markdown vs \~50KB for the full OpenAPI spec. When working with a token-limited AI, prefer `llms.txt` for quick understanding and `openapi.json` only when the agent needs exact schema definitions.