Generate an llms.txt for your site
Paste your domain. We read your sitemap, group your pages into sections and hand you a file to edit and publish at /llms.txt.
Reads your sitemapYours to edit
What it costs
- Here, right now
- Free, no key
- Through the API
- 1 credit a page
- On signup (work email)
- 500 free credits
Reads your sitemap
Finds the pages you already publish and groups them into sections, rather than asking you to list them by hand.
A shortlist, not a dump
An llms.txt is curated by definition, so archives are capped and pagination, tags and account pages are left out.
Yours to edit
Link names come from your URLs, so it is a first draft you correct in a minute, not a file you publish blind.
Questions
A plain markdown file at the root of your site that tells a model which pages are worth reading and how they fit together. It is a convention rather than a standard, and it is advisory: it orients a model that has already reached you, it does not grant or deny access the way robots.txt does.
From the URLs themselves. Reading every page to take its real heading would be one request per link, which no free tool can carry, so the names are a first draft. Fix the ones that read badly before publishing.
Two reasons. Pagination, tag, category and account paths are skipped because they are not worth pointing a model at, and the file is capped overall so it stays a shortlist. Curation is the point of llms.txt.
Serve it at /llms.txt as plain text. Nothing else is required, and nothing breaks if a model ignores it.
No, and anyone promising that is overselling it. It makes a model that already reached you read you accurately. Discovery is a separate and harder problem.
[ Start ]
This is one page. The API does millions.
Same extraction, one credit a page, failures never billed.
Success rate
Median scrape
ms
