Free · runs in your browser

Free llms.txt generator

Fill in a name, one summary sentence and the URLs you want assistants to read. The file is written as you type and never leaves this page.

An llms.txt is a markdown index at your domain root that tells large language models which of your pages are worth reading, in the order you want them read. It is an H1 with the site name, a one-sentence blockquote summary, and H2 sections of markdown links with short notes. It controls orientation, not access — robots.txt still decides whether a crawler may fetch anything at all.

Build your llms.txt

Everything below runs in your browser. The preview updates on every keystroke.

Becomes the file's only H1.

Becomes the blockquote under the H1. Say what the site is and who it is for — no preamble.

A short paragraph or two of orientation. Blank lines start a new paragraph.

llms.txt
  • Add the site or product name — it becomes the file's only H1.
  • Add a one-sentence summary — assistants quote it verbatim.
  • Add at least one URL, or the file tells a crawler nothing it did not already know.

How to create and publish an llms.txt

  1. 1Name the siteEnter the site or product name exactly as you want an assistant to say it. It becomes the file's only H1, and it is the line most likely to be quoted back at a reader.
  2. 2Write one summary sentenceSay what the site is and who it is for, in one sentence with no marketing preamble. It becomes the blockquote directly under the H1 — the first thing a model reads after the name.
  3. 3List the pages you want quotedPaste one URL per line under each heading. Add a title and a short note with pipes: Pricing | https://example.com/pricing | Plans and what each includes.
  4. 4Copy or download the fileTake the generated text and publish it at https://yourdomain.com/llms.txt, served as text/plain. No redirect, no login wall.
  5. 5Verify it is reachableOpen the URL in a private window and confirm it returns the plain file, then check that robots.txt does not disallow it for the AI crawlers you care about.

Frequently asked questions

What is llms.txt?

llms.txt is a markdown file at the root of a website that gives large language models a short, clean index of the pages you want them to read. It was proposed by Jeremy Howard of Answer.AI in 2024. The format is an H1 with the site name, a blockquote summary, and H2 sections containing markdown link lists with optional notes.

Where do I put the llms.txt file?

At the root of the domain, so that it answers at https://yourdomain.com/llms.txt. Serve it as text/plain with a 200 status and no redirect. If you host several country domains, publish one per domain rather than pointing them all at a single file.

Does llms.txt improve my ranking in Google?

No. Google has not said it reads llms.txt, and it is not a ranking signal. What it does is make your key pages cheap to read for the assistants that do fetch it, and give you one canonical place to state what your site is. Treat it as a cheap, low-risk addition, not as a substitute for the pages themselves.

Is llms.txt the same as robots.txt?

No. robots.txt controls access: it tells a crawler which paths it may request. llms.txt controls orientation: it tells a model which of the pages it is allowed to read are the ones worth reading. You need both, and a site can easily publish a perfect llms.txt while robots.txt blocks the crawler that would have fetched it.

What about llms-full.txt?

llms-full.txt is the optional companion that carries the full text of your documentation in one markdown file, rather than links to it. It is useful when your content is documentation-shaped and stable. Start with llms.txt; add llms-full.txt only if you can keep it generated from the same source as the pages, because a stale copy is worse than none.

How many links should I include?

Enough to cover what you want quoted and no more. A good llms.txt for a product site is usually ten to forty links: the product and pricing pages, the core documentation, the company page and the legal pages. Anything you would not want read aloud in an answer does not belong in the file.

Is anything I type here uploaded?

No. The file is assembled by JavaScript running in your browser. Nothing is sent to Search Genie, nothing is logged, and closing the tab discards everything. You can confirm it by opening your browser's network panel while you type.

Can I use this for an Arabic or bilingual site?

Yes. The file is plain UTF-8 markdown, so Arabic headings, summaries and link titles all work. For a bilingual site, list the Arabic pages under their own heading with Arabic titles: an assistant answering an Arabic query matches the Arabic wording, not a translation of your English label.

Is your llms.txt actually reachable?

The free brand audit fetches your robots.txt, sitemap and llms.txt the way a crawler does, samples your pages, and reports what it found — with the URL of every file it read.

What the free audit checks