llms.txt generator · free, no sign-up
Generate an llms.txt for your site
Type your address. We read your sitemap and your pages' own titles and descriptions, and write the curated index that tells an AI assistant what your site contains. Nothing is invented — every line comes from a page you already published.
What makes an llms.txt worth having
The file is short by design. Its value is curation: an assistant that reads it should be able to pick the one page that answers a question, and a file listing every URL on the site gives it no more help than the sitemap already did.
One sentence per page, written for a reader
The description is the whole point. “Pricing” tells an assistant nothing; “Plans and per-article prices, with what each quality tier changes” tells it exactly when to cite that page. If our generator lists a page by title alone, that page has no meta description — fix that first, and both this file and your search snippet improve.
Only the pages you would defend
Cut tag archives, paginated listings, thin category stubs and anything you would not want quoted. A curated forty beats an exhaustive four hundred.
Grouped the way a person would group them
Sections give an assistant a cheap way to narrow before it reads. We group by the first path segment because that is usually how a site is already organised — rename the sections to whatever your visitors would call them.
Kept current
A file generated once and forgotten is worse than none, because it will confidently point at pages that no longer exist. Generate it from your own page data if you can, so it updates when the site does.
Ours is at itmaster.uk/llms.txt if you want to see one in use, and the term is defined in our glossary.
questions people ask
What is llms.txt?
A markdown file at the root of your site — /llms.txt — that lists your most important pages with a sentence about each. It is to AI assistants roughly what a sitemap is to a search crawler: a curated index, so an assistant can find the page that answers a question instead of guessing from links. The convention was proposed in 2024 and is published at llmstxt.org.
Do AI assistants actually read it?
Support is not universal and nobody should claim otherwise. Several coding and research assistants fetch it today, and it costs one static file to be ready for the ones that do. Treat it as cheap insurance and a forcing function for writing a real description on every important page — not as a ranking trick.
Where does the content come from?
Your own site. We read your sitemap (following sitemap index files), take the most important pages by URL depth, and read each page's own title and meta description. Nothing is invented: if a page has no description, it is listed by title alone, and that is a signal worth acting on.
What if I do not have a sitemap?
We fall back to the internal links on your homepage, which gives a usable but much thinner result. If that happens, publishing a sitemap is the higher-value fix — it helps ordinary search crawlers far more than this file helps assistants.
Where do I put the file?
At the root of your domain, so it answers at yoursite.com/llms.txt as plain text. On WordPress you can drop it in the site root or serve it from a plugin; on Next.js, add a route at app/llms.txt/route.ts. Ours is generated from our page data so it cannot go stale, which is the approach we would recommend once the file matters to you.
Should I edit what you generate?
Yes. This gives you a correct, complete starting point in a few seconds; a human should then cut the pages that do not matter and sharpen the descriptions. The convention rewards curation, and a file that lists everything is barely better than no file at all.
A file is the easy part
Being worth citing is the hard part. IT Master writes and publishes the pages that earn citations, and checks every one with models that did not write it.