Short answer: llms.txt is a proposed markdown file at the root of a website that gives AI agents a short, curated overview of the site and links to the detail. Google says it isn’t needed for Google Search and won’t help or hurt your rankings, though other services may use it.
What goes in an llms.txt file
| Part | What it holds | Required |
|---|---|---|
| H1 | The name of the site or project | Yes, the only required part |
| Blockquote | A short summary with the key information | No |
| Details | Paragraphs or lists, without headings | No |
| File lists under H2 | A markdown list of links as [name](url), with optional notes after a colon | No |
| Optional section | Secondary links an agent can skip | No |
From the /llms.txt proposal, v2, by Jeremy Howard, read November 3, 2026.
The file sits at /llms.txt or at any path inside the site, and covers the pages under that path. The proposal adds that its links should point to content language models read easily, such as markdown versions of the pages.
What Google says about llms.txt
Google’s guide to generative AI features lists llms.txt files among the things you can ignore for Google Search. You don’t need new machine-readable files, AI text files or Markdown to appear in Search, including its AI features, because Google Search doesn’t use them.
Keeping one for other services is fine: it won’t harm or help your visibility or rankings, since Google Search ignores it. Google added that note on June 15, 2026.
The same guide says structured data isn’t required for generative AI search either. To see how your pages do in those features, Google points to the Generative AI performance report in Search Console. Our guide to Google AI Overviews covers what a page needs to appear in them.
Who reads an llms.txt file
The proposal’s author says thousands of sites publish one, documentation platforms generate it automatically, and Chrome’s Lighthouse audits sites for one as part of its agentic browsing checks.
OpenAI, Anthropic and Gemini publish llms.txt files for their own developer docs. The files are used most for software documentation, though a business can use one to outline its structure and policies.
It doesn’t replace robots.txt or your sitemap. Robots.txt says what access is acceptable, a sitemap lists every page for search engines, and llms.txt offers a curated overview for language models.
How to write an llms.txt file
- Choose the pages an AI agent most needs to understand your business, such as services, prices and contact.
- Start with an H1 that holds the name of your business.
- Add a blockquote with a short summary of what you do and for whom.
- Group your key pages under H2 sections, as a markdown list with one link per item.
- After a link, add a colon and a short note where it helps.
- Put secondary pages under a section called Optional.
- Publish it at /llms.txt and update it when key pages change.
Our guides to robots.txt, the XML sitemap and structured data cover the files Google does use.
What our own llms.txt holds
Our own llms.txt, read on November 3, has 197 lines and 21,357 bytes. It opens with an H1 with our name and a one-paragraph summary in a blockquote, as the proposal asks.
Below come 8 H2 sections, from Services to Contact, with 161 links: 91 of them to our English guides, one per guide. We add a line each time a guide goes live, and our server keeps a copy from before each change: 72 so far.
One gap against the proposal: our links point to the normal HTML pages, not to markdown versions. We keep the file for AI agents and other services, not for Google Search.
Before you publish an llms.txt file
- The file starts with an H1 with your name.
- A short summary sits in a blockquote.
- Key pages are grouped under H2 sections.
- Every link is a markdown link that works.
- Secondary links sit under Optional.
- It opens as plain text at /llms.txt on your domain.
RememberAn llms.txt file can help AI agents find their way around your site, but Google Search doesn’t use it. Keep it short, update it when a key page changes, and put your SEO work into the pages it links to.
If you want your pages to be the ones AI answers cite, see our AI search optimization service.
Quick answers about llms.txt
Will Google crawl my llms.txt file?
It may. Google says it can discover, crawl and index many kinds of files besides HTML, which doesn’t mean it treats them in a special way. Anything in the file is public, so write it as you would a page anyone can read.
Is llms.txt the same as robots.txt?
No. Robots.txt tells automated tools what access to a site is acceptable. An llms.txt file gives language models a curated overview of the site, with links to the pages where the detail lives.
Where does the llms.txt file go?
At the root of the site, as /llms.txt, or at any path inside it, where it covers the pages under that path. When more than one file applies, the proposal says agents should use the most specific one.
What format does llms.txt use?
Markdown. The only required part is an H1 with the name of the site. A short summary in a blockquote, details without headings and lists of links under H2 sections can follow, in that order.
Should the links point to markdown versions of my pages?
The proposal says the links should point to content that language models read easily, such as clean markdown versions of the pages at the same address, as page.html.md or page.md. Plain HTML links still work, as ours do.
Sources
Read on November 3, 2026. The /llms.txt file, v2 (llmstxt.org). Google Search Central: Optimizing your website for generative AI features on Google Search and Latest documentation updates. Our own case: our /llms.txt file, November 3, 2026.