Free tool · built from your site

llms.txt generator A first draft, made from what your site already says

We read your title, your description and your sitemap, fetch the real page titles, and give you a file to edit and upload.

Free. This one reads a few of your pages, so give it a moment.

Where each line comes from

The name line
The # name is your homepage <title>.
The summary line
The > summary is your homepage meta description.
The links
The links are your sitemap, with each page's real title fetched.
Anything unreadable
Anything we could not read comes back in capitals for you to replace.

An honest word about this file

What we have not seen
No evidence that it changes anything. No major engine has committed to reading llms.txt, and nothing in our scan data suggests having one changes whether a site gets cited. Plenty of tools will tell you otherwise. We would rather you knew before you spent the afternoon.
Why it is still worth doing
Ten minutes, and nothing to lose. It is still worth ten minutes. It costs nothing, it is trivial to add, and if the convention is adopted you are already done. Just do the things that the full check scores first.

Questions about llms.txt

What is an llms.txt file?

A plain-text file at your site root that tells AI assistants what the site is and points at the pages worth reading. It is a proposed convention rather than a standard anyone has adopted, which is worth knowing before you spend an afternoon on it.

Do AI engines actually read llms.txt?

No major engine has said it does, and we have seen nothing in our own scan data suggesting a site with one gets cited more. We would rather tell you that than sell you a file. It costs ten minutes, it cannot hurt, and it is not the thing standing between you and being quoted.

How do I create an llms.txt file?

Put your address in above. We read your homepage title and description and your sitemap, fetch the titles of the first pages in it, and give you the file. Save it as llms.txt at your site root so it answers at /llms.txt.

What should be in it?

An # line naming the site, a > line saying in one sentence what it is for, then Markdown links to the pages that matter. Short and specific beats long and complete: it is a guide, not a sitemap.

Why are some of the titles wrong?

Because we used what the page said. Where a page did not answer in time we used its URL path instead and flagged it, rather than inventing a name for it. Anything the file marks in capitals is a gap you need to fill yourself.

Is this the same as the llms.txt checker?

The other way round. The checker reads a file you already have and tells you whether it is valid. This writes a first draft for a site that has none.

Already have one?

The checker reads yours against the proposed format and tells you what is wrong with it.

Check my llms.txt