What is llms.txt and should your site have one?
llms.txt is a proposed convention — published at llmstxt.org — for a markdown file at your site's root that gives LLMs a curated map of your most important content: what the site is, in one blockquote, followed by organized links with one-line descriptions. Where robots.txt tells crawlers what they may fetch, llms.txt tells language models what is worth reading and how it fits together. Adoption by the engines is still uneven, so treat it as cheap insurance with a real secondary benefit: writing one forces you to decide which pages you actually want quoted, which is the AEO exercise in miniature.
The format is deliberately simple: an H1 with the site name, a blockquote summary, then sections of links. The spec's design goal is a file small enough to fit in a context window while pointing at everything that matters — which is exactly what you want an engine to see first.
Generate it from the same source as your pages rather than writing it by hand, so it cannot drift: a new answer page should appear in llms.txt automatically. Ours works this way, and drift-proofing is the difference between a live map and a stale one.
Keep expectations calibrated: no major engine currently commits to consuming llms.txt, and it is not a ranking lever on its own. It costs an hour, it cannot hurt, and if the convention wins you were early. That is the honest case for it — not more.
Sources
Last reviewed 2026-08-10. We revisit these when the underlying research changes.
Related
See where you actually stand
Run a free check across ChatGPT, Perplexity, Gemini, Claude and Google AI Mode.
Check your AI visibility