Skip to content
AI Search (AEO & GEO) 4 min read Sajid Aslam

llms.txt Explained: Does Your Site Need One?

llms.txt is a sensible idea for documentation sites. For most business websites it is optional, and Google says it ignores it.

Example structure of an llms.txt file with headings and links

Short answer

llms.txt is a proposed Markdown file at the root of a website that gives AI tools a short summary of the site and links to its most useful pages. It was proposed by Jeremy Howard in September 2024. Google says Google Search ignores it. It is useful for documentation sites read by AI coding tools; for most business sites it is optional.

llms.txt is a proposed file, written in Markdown and placed at the root of a website, that gives AI tools a short description of the site and a curated list of links to its most useful content. It is a sensible idea for some sites. It is not a ranking factor, Google says its Search ignores it, and for most small business websites it is optional and low priority.

This article covers what the file is, who actually reads it, and when it is worth ten minutes of your time. For the wider set of things that do affect AI visibility, start with the AI search optimisation guide.

What llms.txt is

The llms.txt proposal was published by Jeremy Howard on 3 September 2024, and updated to a second version in August 2026. The problem it addresses is real: web pages are full of navigation, scripts and layout that make them awkward for a language model to read within a limited context, especially for something like software documentation. A short Markdown file pointing to clean versions of the key pages is easier to use.

The proposal describes a file at /llms.txt containing, in order:

  1. An H1 with the name of the site or project — the only required part.
  2. A blockquote with a short summary.
  3. Optional paragraphs or lists with more detail.
  4. Sections under H2 headings, each a list of links with a short description.
  5. Optionally, a section called "Optional" for links that can be skipped when space is short.

It also suggests publishing clean Markdown versions of pages, at the same URL with .md added.

Who actually reads llms.txt

This is where claims and evidence part company.

ReaderUses llms.txt?Basis
Google Search, AI Overviews, AI ModeNoGoogle's AI optimisation guide says Search ignores it
ChatGPT searchNot documentedOpenAI's crawler docs cover robots.txt only
PerplexityNot documentedPerplexity's crawler docs cover robots.txt only
AI coding assistants and agentsSometimesDeveloper tools fetch it when pointed at documentation
Chrome LighthouseChecks itAn agentic browsing category added in 2026 includes an llms.txt check (release notes)

Google's guide, at the time of writing (October 2026), says having an llms.txt file will neither harm nor help your visibility or rankings in Google Search, and that it is fine to keep one for other services that use it. Before that guidance was written, Google's John Mueller had compared the file to the old keywords meta tag.

The Lighthouse check is worth understanding correctly. Lighthouse is Google's site-auditing tool, not Google Search. Its check looks at whether a file you publish is well-formed. It is not a signal that Search uses the file, and Google's Search documentation says otherwise.

Does your site need an llms.txt file?

Site typeRecommendation
Software product with documentation, an API or a developer audienceYes, worth doing properly
SaaS or tool where people ask AI assistants how to use itProbably worth it
Local service business, brochure site, small shopOptional; do it only after the basics
Any site with crawlers blocked in robots.txtFix robots.txt first; llms.txt changes nothing

For a plumber, an accountant or a small online shop, the things that affect AI visibility are crawler access, indexing, clear pages and consistent business information. How to get cited by ChatGPT and Perplexity and how to appear in Google AI Overviews cover those. An llms.txt file sits well below them.

If you do create one

Keep it short, accurate and maintained.

  1. Start with the business name as the H1 and a one- or two-sentence summary in a blockquote: what you do, where, for whom.
  2. Add an H2 section for services, with a link to each service page and a one-line description.
  3. Add an H2 section for your most useful guides — the pages that answer real customer questions.
  4. Add contact and About links.
  5. Put lower-priority links under an "Optional" heading.
  6. Check it whenever you add, remove or rename pages.

Do not paste your whole sitemap into it; that defeats the purpose. Do not put claims in it that are not on the site. And do not expect anything from Google as a result.

Common llms.txt claims, checked

ClaimWhat the evidence says
"llms.txt helps you rank in Google"Google says Search ignores it and it neither helps nor harms
"ChatGPT and Perplexity prioritise sites with llms.txt"Neither documents using it for search; their docs cover crawlers and robots.txt
"Lighthouse checks it, so Google must use it"Lighthouse audits sites; Google Search is a separate system with its own documentation
"It stops AI training on your content"No — it grants and blocks nothing; use robots.txt rules for training crawlers
"AI labs use it themselves"True for their developer documentation — which is the use case it was designed for

The risk worth naming

The real cost of llms.txt for a small business is not the file. It is being sold it. If an agency charges for "AI optimisation" and the deliverable is an llms.txt file, you have paid for something Google says it ignores and that the major AI search products do not document using. Ask what else is in the work.

Next step

If you have ten minutes for AI search this week, spend them on robots.txt and how Google crawls and indexes your site, not llms.txt. If you want both checked properly, the SEO service covers crawler access as part of the audit.

Related services

Related reading

FAQ

Questions about this

If yours isn't here, send it over — I reply within one working day.

No. robots.txt tells crawlers what they may and may not fetch, and the major crawlers obey it. llms.txt is a voluntary summary that points AI tools to useful content; it grants or blocks nothing. If you want to control AI crawlers, robots.txt and your firewall are the tools.

No. Google says it neither helps nor harms visibility in Search because Search ignores it. The only risk is maintenance: a file listing pages that no longer exist, or describing the business inaccurately, is worse than none for any tool that does read it.

OpenAI does not document using llms.txt for ChatGPT search; its documentation covers crawlers and robots.txt. An AI tool may read the file if a user points it there, as it would any page. Treat claims that ChatGPT prioritises sites with llms.txt as unsupported.