What is llms.txt?

Jeremy Howard proposed llms.txt on 3 September 2024. The proposal page was last changed on 10 August 2026 (llmstxt.org, read 1 November 2026). It calls itself "A proposal to standardise on using an /llms.txt file to provide information to help agents use a website."

No standards body has adopted it. Unlike robots.txt, which is an IETF standard, llms.txt is a community proposal.

What goes in an llms.txt file?

  1. An H1 with the name of the site or project. This is the only required part.
  2. A blockquote with a short summary.
  3. Optional paragraphs or lists with more detail (no headings).
  4. H2 sections with lists of links, each with an optional note.
  5. An "Optional" section for secondary links that tools can skip.

The proposal also suggests a clean Markdown copy of important pages at the same URL with .md added. A file can sit at the root or in a folder, such as /docs/llms.txt for the pages under /docs/.

Does llms.txt help you rank or get cited?

Not in Google. Google's AI features page says: "You don't need to create new machine readable files, AI text files, or markup to appear in these features" (Google Search Central, last updated 10 December 2025).

We found no page from OpenAI, Anthropic or Perplexity saying their search products read llms.txt. Some AI companies publish one for their own docs: Anthropic has a long llms.txt for its developer docs. Publishing a file is not the same as using other sites' files to pick answers.

So treat it as cheap hygiene, not a lever. It can help coding tools and agents find your docs. It will not make an assistant name your brand.

Why does llms.txt matter for B2B SaaS?

Products with API docs or help centers gain most, because developers paste those docs into AI tools. For a marketing site, a short file that states what you sell, for whom, and links to pricing and docs is enough. Keep it in line with your site: a wrong llms.txt is worse than none.

What does a file look like? (hypothetical example)

This example is hypothetical, for a help desk company:

llms.txt vs robots.txt vs structured data

  • robots.txt tells crawlers what they may fetch. Its standard (RFC 9309) says the rules "are not a form of access authorization" (RFC 9309).
  • llms.txt suggests what an AI tool should read first. It allows or blocks nothing.
  • Structured data labels facts inside a page, such as a product's price.
  • An AI crawler is the bot that may or may not read any of these files.

More on what does move AI visibility: LLM SEO: how B2B SaaS brands get named in AI answers.