# llms.txt explained: the standard that makes AI understand your website

> **llms.txt** is a Markdown file in a website's root directory (at `/llms.txt`) that gives language models a curated, compact map of the most important content. Proposed in September 2024 by Jeremy Howard (Answer.AI). It answers, for the machine, in one file: What is this about, and what should I read first?

Machine Web · Reference · Updated: 2026-07-03

An ordinary website is built for the human eye. A language model has to laboriously extract the meaning from it. `llms.txt` turns that around: instead of letting the machine guess, you hand it a deliberately curated summary — concise, ordered, in plain Markdown. It does not replace the page's content, it makes it accessible.

## Context: robots.txt, sitemap.xml, llms.txt

All three files live in the root directory and address machines — each answers a different question.

| File | Answers | Role |
|---|---|---|
| `robots.txt` | May I read this? | controls **access** |
| `sitemap.xml` | Which addresses exist? | lists the **URLs** |
| `llms.txt` | What is it about, what matters? | provides **understanding** |

`robots.txt` and `sitemap.xml` tell a machine *what* it may fetch and *where* it is. `llms.txt` adds the missing layer: *what it means*.

## How an llms.txt is structured

The format is deliberately simple: valid Markdown, readable in seconds, trivial to parse. The order is part of the convention:

1. **H1 with the name** of the website or project.
2. **Blockquote with a short summary** (`>`) directly below it.
3. **Optional explanatory paragraphs** as ordinary running text.
4. **H2 sections with annotated links** in the format `Name: Description`.
5. **An "Optional" section** for secondary material that may be skipped.

### Small example

```markdown
# Example Project

> A small tool that translates Markdown into clean HTML pages.
> Open source, intended for developers and automation.

## Documentation

- [Quickstart](https://example.com/docs/start): Installation in five minutes.
- [API reference](https://example.com/docs/api): All functions with parameters.

## Optional

- [Changelog](https://example.com/changelog): Version history — mostly not needed.
```

The strength lies in the curation, not in the length.

## Creating an llms.txt — step by step

1. **Create the file** as `llms.txt`, reachable at `https://your-domain/llms.txt`.
2. **Write the header:** H1 (name) plus a blockquote with a short summary.
3. **Curate the content:** only the pages a machine really needs, grouped into H2 sections.
4. **Annotate the links:** one sentence each in the format `Name: Description`.
5. **Offload the secondary** into an "Optional" section.
6. **Optionally add `llms-full.txt`** for the full text in a single fetch.
7. **Check:** open it in the browser — it should appear as plain Markdown text.

## llms-full.txt — the full text in one file

Where `llms.txt` is a *map*, `llms-full.txt` is the *whole book*: the complete textual content of the website in a single file. This way a model gets the complete content without further fetches. The two complement each other — the map for the overview, the full text for the exhaustive answer.

## Status & limits

`llms.txt` is **not an official W3C standard**, but an open convention (2024) with growing adoption.

- Not every language model evaluates `llms.txt` yet today.
- It complements `robots.txt`, `sitemap.xml` and structured data (JSON-LD) — it replaces none of these building blocks.
- The value comes from curation, not from automatic generation.

## Frequently asked questions

**What is llms.txt?** A Markdown file at `/llms.txt` with H1, blockquote and H2 sections of annotated links — curated AI context. Proposed in 2024 by Jeremy Howard (Answer.AI).

**What is llms.txt good for?** It gives language models understanding: a compact, curated map of the website. robots.txt controls access, sitemap.xml lists URLs, llms.txt provides meaning.

**How do I create an llms.txt?** A Markdown file in the root directory: H1 (name), blockquote (summary), H2 sections with links in the format "Name: Description", plus an "Optional" section.

**Difference between llms.txt and llms-full.txt?** llms.txt is the short, curated map; llms-full.txt contains the full text of the entire website in one file.

**Is llms.txt an official standard?** No — not a W3C standard, but a format proposed in 2024 with growing adoption.

**Does llms.txt replace robots.txt/sitemap/structured data?** No, it complements them.

## How KAMINSKI+ leads by example

KAMINSKI+ serves in production a maintained [/llms.txt](https://kaminskiplus.de/en/llms.txt) and the complete [/llms-full.txt](https://kaminskiplus.de/llms-full.txt). The [Guide for machines](https://kaminskiplus.de/en/leitfaden/) shows the interplay with the sitemap, API and grounding pages; the reference [AI-readable website](https://kaminskiplus.de/en/machine-web/ai-readable-website/) places llms.txt in the larger picture of a machine-first built site.

## Further reading

- [Machine Web (overview)](https://kaminskiplus.de/en/machine-web/) — all reference topics
- [AI-readable website](https://kaminskiplus.de/en/machine-web/ai-readable-website/) — building machine-first
- [Guide for machines](https://kaminskiplus.de/en/leitfaden/) — the operating manual
- [Ground truths](https://kaminskiplus.de/en/facts/) — verifiable facts without hallucination