What Is llms.txt?
llms.txt is a proposed standard file that gives AI models a clean, curated guide to your site's most important content. Definition, what goes in it, and whether it helps.
llms.txt is a proposed web standard: a plain Markdown file placed at the root of a domain (for example yoursite.com/llms.txt) that gives large language models a clean, curated guide to your most important pages and facts. It works like a sitemap written for AI models rather than search crawlers, pointing an assistant at the content you most want it to read when it reasons about your site. It is an emerging convention, not a ratified standard or a confirmed ranking factor.
- llms.txt is a Markdown file at your domain root that curates your key pages and facts for AI models, much like a sitemap aimed at assistants rather than crawlers.
- It is a proposed convention, not an official standard, and no major AI provider has confirmed it changes how you are cited.
- Because it is cheap to add and easy to maintain, it is a low-risk experiment rather than a guaranteed win.
- It complements the real work of being cited by trusted third-party sources; it does not replace it.
What goes into an llms.txt file
The format is deliberately simple, which is much of the point. A file typically opens with an H1 carrying your site or product name, followed by a short blockquote that summarizes what you do in a sentence or two. Below that sit one or more sections, each an H2 grouping a handful of Markdown links, and each link paired with a short description of what the reader will find there. The goal is a curated map a model can parse in seconds, not an exhaustive dump of every URL.
- A title and a one-line summary that states plainly what your company or product is.
- Links to your most important pages: core documentation, product and pricing, key guides, and an about or company page.
- A short, factual description beside each link, so the model knows why the page matters and when to use it.
- An optional secondary section for lower-priority links, keeping the primary list tight and signal-rich.
Some sites also publish a companion llms-full.txt that inlines the full text of those pages in one file, so a model can ingest the content directly without following each link. Keep either file short and current: the value is in the curation, so treat it like the briefing you would hand a new analyst on day one, not a mirror of your whole site. The proposal and its formatting rules are documented at llmstxt.org.
Does llms.txt actually help?
Honestly, the evidence is still thin. Adoption is uneven: a number of developer-tool and documentation-heavy companies now publish an llms.txt file, but the major AI labs have not publicly committed to reading it, either at training time or when an assistant answers a live question. There is no confirmed, measured lift in citations from adding one, so any claim that llms.txt is a ranking factor should be treated skeptically for now.
The case for adding one anyway is that the cost is close to zero and the exercise is useful on its own terms. Writing the file forces you to name your most important pages and state your core facts plainly, which is good hygiene whether or not a model ever reads it. If adoption does grow, an early, well-maintained file costs little to keep current, and you are positioned rather than scrambling.
What it will not do is manufacture credibility. AI answers are assembled from the third-party sources an engine already trusts, and a file you publish about yourself cannot substitute for being named on those sources. Getting cited in answers still comes from earning a presence on the review sites, comparisons and reference pages engines lean on, the discipline of answer engine optimization (AEO). Treat llms.txt as one small, optional signal layered on top of that work, not a shortcut around it.
An llms.txt file is a reasonable, low-effort experiment for most sites. Just set expectations accordingly: it may make your site easier for a model to understand, but it will not move your AI search visibility on its own. Measure that separately, and let the data tell you whether anything changed.
How to add and check one
Adding the file is a one-time job: write the Markdown, save it as llms.txt, and serve it at the root of your domain so it resolves at yoursite.com/llms.txt. Before you invest much in it, though, confirm the more basic thing: that AI crawlers are actually allowed to reach your site in the first place. A restrictive robots.txt or a missing llms.txt is easy to overlook and undoes the effort. Our free AI crawler checker inspects both files and tells you which AI bots your site currently welcomes or blocks, so you can fix the access problem before worrying about curation.
Frequently asked questions
What is llms.txt?
llms.txt is a proposed plain-text file placed at your site root that gives AI models a curated, markdown-formatted guide to your most important pages and facts, so they can find and understand your content without crawling everything. It is to AI models roughly what a sitemap is to search crawlers.
Does llms.txt actually help AI visibility?
It is an emerging convention, not a guaranteed ranking factor, and adoption by engines is still evolving. It is low-cost to add and helps ensure your key facts are available in a clean form, but it complements, rather than replaces, being well cited across the wider web.
We build Measure, a self-serve platform that tracks how AI engines describe and recommend brands, and connects that visibility to real traffic and revenue. Everything here is written from what we see in the data every day.