llms.txt is a proposed plain-text file at the root of a site that tells AI models what the site contains and where its most useful pages are.
llms.txt is a proposed convention: a plain-text Markdown file at the root of a
domain that describes what the site is and links to its most useful pages. The
idea is roughly robots.txt in reverse. Where robots.txt restricts what
crawlers may fetch, llms.txt advertises what a model would most benefit from
reading.
A short, consistent structure:
# H1 with the site or product name.> blockquote giving a one-line description.## sections listing pages as Markdown links, each with a brief
description of what it covers.The format is deliberately plain so that a model can consume it without parsing HTML or executing JavaScript.
Not for Google. Google's guide to its AI features lists "LLMS.txt files and other 'special' markup" among the things you do not need to appear in AI Overviews or AI Mode. No other major AI provider has confirmed that it reads the file.
The argument for publishing one anyway is cost. It is one file, it can be generated from the same source as your sitemap so it never goes stale, and it does no harm. Spend the time on pages that answer their search instead.
A sitemap is a machine-readable list of every URL, aimed at search crawlers, and
it carries no meaning beyond the addresses and when they changed. An llms.txt
is a short curated description aimed at a reader, and it says what each page is
for.
A sitemap answers "what exists here". An llms.txt answers "what is this site,
and what should I read first".
What does help is in what does and doesn't affect Google rankings. See also answer engine optimization.