# For agents and developers

> Machine-readable ways into farez.ai: llms.txt, Markdown for every page via Accept: text/markdown, a sitemap, RSS and a JSON index of the Field Guide.

Web page: https://www.farez.ai/developers

Everything on farez.ai is public and read-only: there are no accounts, API keys or rate-limited tiers. These are the ways in for AI agents, scripts and tools.

## Quickstart

```sh
# The curated index for agents (what's here, and when to use it)
curl https://www.farez.ai/llms.txt

# Any page as Markdown: ask for it in the Accept header…
curl -H 'Accept: text/markdown' https://www.farez.ai/guide/transformer/

# …or add .md to the path (the homepage is /index.md)
curl https://www.farez.ai/guide/transformer.md
```

## What's available

| Resource | URL | What it is |
|---|---|---|
| llms.txt | [/llms.txt](/llms.txt) | Curated index in the [llms.txt](https://llmstxt.org) format, including when an agent should use this site |
| Markdown pages | `Accept: text/markdown` or `<path>.md` | Every lesson, concept, paper and post as Markdown (`Content-Type: text/markdown`, `Vary: Accept`) |
| Sitemap | [/sitemap-index.xml](/sitemap-index.xml) | Every page |
| RSS | [/rss.xml](/rss.xml) | Blog posts |
| Field Guide index | [/guide/search.json](/guide/search.json) | Every AI term: slug, name, region, tier, one-line definition |

Unknown paths return a real `404`. Ask for Markdown and the 404 body is Markdown too, pointing back here.

## Using the content

Linking to and quoting from the site with attribution is welcome. Facts on the site cite their primary sources; when you rely on a fact, prefer the source it cites. Found something wrong, or want to use the content in another way? [Get in touch](/contact).

Last updated: 2026-10-05