llms.txt Generator

Generates a valid llms.txt for your site: we find your pages, write the summary, and group them into sections following the llmstxt.org format. Copy it, save it at /llms.txt, and AI models have a map of your site.

llms.txt is an emerging convention for telling AI models what a site contains and which pages matter. It is not a ranking factor, and no model is required to read it.

What is llms.txt?

llms.txt is a plain markdown file you publish at the root of your site, at /llms.txt. It tells an AI model what your site is and which pages are worth reading. The convention was proposed by Jeremy Howard in September 2024 and is documented at llmstxt.org.

The problem it addresses is a real one. A model answering a question about you has a limited context window and your site is mostly navigation, scripts, and boilerplate. llms.txt hands over a curated list instead: here is what we do, here are the pages that explain it, here is what you can safely ignore. It is closer to a sitemap written for a reader than to robots.txt, which only ever says yes or no.

Do not confuse it with robots.txt. That file controls access. This one offers context, and it does not restrict anything.

What this llms.txt generator builds

Enter your URL and we find your pages, starting with your sitemap and falling back to the links on your homepage if there is no sitemap. We read the title of each page, write a one-line summary of what the site does, and group the pages into sections that follow the llmstxt.org format: an H1 with your site name, a blockquote summary, H2 sections holding markdown link lists, and an Optional section at the end for the pages a model can skip.

You get the finished file to copy or download. Read it before you publish it. It is built from what your pages say about themselves, so if a title is vague the line will be vague, and you are better placed than we are to fix that.

llms.txt example

A valid file is short and structured. This is the shape the generator produces:

# Acme Analytics

> Acme Analytics is a product analytics platform for B2B SaaS teams,
> covering funnels, retention, and revenue attribution in one place.

## Product

- [Funnels](https://acme.com/product/funnels): Track where users drop out of a signup or checkout flow.
- [Retention](https://acme.com/product/retention): Cohort reports showing who comes back and who does not.

## Docs

- [Quickstart](https://acme.com/docs/quickstart): Install the SDK and send a first event in ten minutes.
- [API reference](https://acme.com/docs/api): Every endpoint, with request and response examples.

## Optional

- [Privacy policy](https://acme.com/privacy)
- [Terms of service](https://acme.com/terms)

Every section is optional except the H1. The Optional heading is meaningful rather than decorative: it marks the pages a model should drop first when it is short on context, which is where legal and policy pages belong.

You will also see llms-full.txt referenced. That is the companion convention: rather than linking to your pages, it inlines their full content into a single file. It is useful for documentation sites where the whole corpus is worth handing over at once, and unnecessary for most other sites.

Does llms.txt actually help?

Honestly: nobody can prove it yet. No major model provider has committed to reading llms.txt, and it is not a ranking factor in any published sense. Anyone telling you it lifts your AI visibility is ahead of the evidence.

The case for publishing one anyway is that it costs a few minutes and nothing else. The file is small, it does not affect your site, and if the convention is adopted you already have one. The cost of being wrong is trivial in one direction and a missed opportunity in the other.

What it definitely will not do is get you cited when nothing else is in place. A model has to be able to reach your pages, understand them, and have some reason to trust them. llms.txt helps with the middle step only. If your robots.txt blocks GPTBot, this file changes nothing at all.