What llms.txt is, and whether it actually helps AI find you
A plain-text map for AI crawlers. What it does, what it does not, and an honest read on whether to bother.
llms.txt is a plain-text file you place at the root of your site that gives AI crawlers a curated map of your most important pages, in Markdown. Think of it as a table of contents written for models rather than people, the way robots.txt is written for search crawlers.
It is a proposed convention, not an official standard, and no assistant has confirmed it as a ranking factor. So the honest question is not just what it is, but whether it earns its place. Here is our take, from actually running one.
What the file contains
An llms.txt file starts with your site name and a short description of what you do, then lists your key pages as Markdown links grouped under headings, each with a one-line summary. The goal is that a model reading it can understand your site and find your best content without crawling every page. Some sites also publish llms-full.txt, a larger file with the full text of key pages inlined.
What it is meant to solve
Models work within a limited context window, and crawling a full site is noisy and expensive. llms.txt offers a clean, human-curated shortcut: here is who we are, here are the pages that matter, described in our own words. In theory that helps a model represent you accurately instead of inferring you from whatever it happened to crawl.
The honest caveat
No major AI platform has publicly committed to reading llms.txt, and it is not a substitute for the things that clearly do matter: server-rendered content, allowing AI crawlers, structured data, and being genuinely citable. If your pages are a mess, a tidy index of them will not save you. Treat llms.txt as a low-cost supplement, not a growth lever.
Why we publish one anyway
The cost is close to zero and the downside is nil. It takes an afternoon to write, it doubles as a clear statement of what your site is about, and if the convention gains adoption you are already there. We publish one for our own site and for clients, kept in sync with the pages we most want represented in AI answers. It is insurance, not strategy.
How to write a useful one
Keep it curated, not exhaustive: list the pages you want a model to represent you by, not every URL. Describe each in plain language, the same answer-first way you would write a quotable passage. Keep it current, because a stale map is worse than none. And serve it as text at yoursite.com/llms.txt so a crawler can actually read it.
Questions
What is an llms.txt file?
It is a plain-text, Markdown file at your site root that gives AI crawlers a curated map of your most important pages with short descriptions, like a table of contents written for models.
Does llms.txt improve AI search ranking?
No AI platform has confirmed it as a ranking factor. It is a low-cost supplement, not a growth lever, and it does not replace server-rendered content, structured data, and genuine citability.
Is llms.txt worth adding to my site?
For most sites, yes, because it costs almost nothing and doubles as a clear statement of your content. Treat it as insurance in case the convention gains adoption, not as a strategy on its own.
SEO ranks a page for a person to click. GEO gets a passage quoted by a model. What actually differs, and what to do about it.
A practical checklist for becoming the source an assistant quotes, platform by platform.
Search traffic increasingly arrives pre-answered. What it takes to be the source an assistant cites, not the tenth blue link.