Free tool
llms.txt
generator.
Enter your site. We read its sitemap and pages and draft the file, then you edit it until it is right.
This tool drafts an llms.txt file for your site. It reads robots.txt and your sitemap, samples up to 40 pages for titles and descriptions, sorts them into sections and puts key pages first. You edit the result and download it. llms.txt is a proposal from llmstxt.org, and not all AI tools use it.
We fetch your robots.txt, sitemap and pages using the name KookiesAEOCheck. We do not store the URL or the results.
Reading your sitemap and pages. This can take up to 10 seconds...
Edit your file
Changes show in the preview right away.
A rough draft from the HTML we read, with pages cut short. It is not part of the llmstxt.org format, so read it through before you use it.
llms.txt is a proposal, not a web standard. Some AI tools read it and some do not, so treat it as a useful extra, not a way to get cited.
How it works
From sitemap
to file.
Reads robots.txt, then your sitemap, including index files. With no sitemap, it uses the links on your home page.
Up to 40 pages for titles and descriptions. It skips noindex pages, tag pages, files and duplicates, and tells you what it left out.
Groups pages by address into sections like Company, Services, Pricing, Case studies, Blog and Docs. Key pages go first, legal pages go in Optional.
Rename, reorder or drop sections, edit each link and note, and watch the file update. The draft is a start, so read it before you publish.
The format
What goes
in the file.
The name of your site. The only required part.
A short summary of what the site is.
Optional plain text with more context. No headings.
Lists of links, each with an optional note after a colon.
An H2 by that name holds links a tool can skip when it needs a shorter read.
Publish it at /llms.txt, at the root of your site.
llms.txt is a proposal, not a web standard. Some AI tools read it and some do not, so treat it as a useful extra, not a way to get cited.
FAQs
llms.txt
questions.

A Markdown file at the root of a site that points AI tools to your key pages, with a short summary. It was proposed at llmstxt.org.
Some do and some do not. It is a proposal, not a standard, and no tool has to read it. It takes little time to add, so many sites treat it as a useful extra. Do not expect it to get you cited on its own.
One check has limits on pages and time, so we read up to 40 pages. We pick the top pages from each section first. Others are listed turned off, titled from their address, so you can turn on and edit what you need.
A longer companion file some sites publish, with page text in one place. It is not part of the llmstxt.org format. We offer a rough draft from the HTML we read. Pages are cut short, so read it through before you use it.
No. robots.txt tells crawlers what they may fetch. llms.txt suggests what is worth reading. Use our AI crawler checker for robots.txt.
We read plain XML sitemaps over https, not compressed .gz files, up to 3 MB each and 5 files from an index. Sites that block bots may also refuse us. You can always start with a blank file.
No. We fetch your pages, build the draft and send it to your browser. We keep your IP address in memory for a short time to limit checks, and we do not save the URL or the results.


