llms.txt
llms.txt is a proposed web standard: a plain markdown file at the root of a website that gives AI systems a curated map of the site's most important content. It tells language models what the site is, what it offers, and which pages matter, in a format built for machine reading rather than human browsing.
What is llms.txt?
llms.txt is a convention modelled loosely on robots.txt, but with the opposite intent: instead of telling crawlers what to avoid, it tells AI systems what to read. The file sits at the site root and contains a short description of the site plus a structured list of its key pages with one-line summaries. The aim is to save models from parsing navigation, scripts and boilerplate to work out what a site actually says. It is a community proposal, not a ratified standard, and public confirmation of how much weight individual engines give it remains limited.
How it works
The file is written in plain markdown: a title, a one-paragraph description of the site, then sections listing important URLs with a short summary of each. For a commercial site, that typically means the product or service pages, pricing context, documentation, and the pages that answer buying questions. It costs little to create and maintain, which is why many teams treat it as cheap insurance: a clear self-description available to any system that chooses to read it, kept current as the site changes.
Why it matters
llms.txt is one of the lowest-effort GEO inputs available: an afternoon of work that makes a site legible to any AI system that looks. drio agency includes it in the technical layer of an engagement, with the honest caveat that it is a supporting signal, not a lever that moves answers on its own.
Key takeaways
- llms.txt is a plain markdown file at a site's root that gives AI systems a curated summary and map of key pages.
- It is a community proposal rather than a ratified standard, and engine support is uneven and largely unconfirmed.
- It is cheap to create and maintain, which makes it sensible low-cost hygiene rather than a primary GEO lever.
Frequently asked
- Does llms.txt actually improve AI visibility?
- llms.txt is best treated as a supporting signal rather than a proven ranking lever. It is a community proposal, and the major engines have not publicly committed to reading it or weighting it. What it reliably does is remove ambiguity: any system that does read it gets an accurate, current description of the site and its key pages. Given the negligible cost, most teams publish one and spend their real effort on content and citations.
- What should an llms.txt file contain?
- A good llms.txt file contains a one-line title, a short paragraph describing what the site and organisation do, and a structured markdown list of the site's most important pages, each with a one-sentence summary. For a commercial site that means the pages that answer buying questions: what the product or service is, who it is for, and how engagement works. It should be kept short, current and factual, because its whole value is being a trustworthy shortcut.
Related terms
Want this working for your brand? Book an intro call.