llms.txt

llms.txt is a proposed plain-text file at a site’s root that summarises the site and links its key pages in Markdown, intended for large language models rather than people. The proposal defines a simple format: an H1 with the site name, a blockquote summary, then H2 sections of links with short descriptions. A companion llms-full.txt can carry the full text of the key pages.

Status

It is a community proposal, not a standard. At the time of writing, none of the crawler documentation we track from OpenAI, Anthropic or Perplexity mentions the file, and Google’s guide to generative AI features says Google Search does not use it. We set out the detail in Does llms.txt work? We publish one as a labelled experiment, generated from the same data as the pages so it cannot drift, and we make no claim that it affects visibility.

Why publish one anyway

It costs little, it is a useful public statement of what the site is about, and it is easy to generate from typed content. The risk is maintaining a second copy by hand; generate it at build time instead.

Our llms.txt tool fetches a site’s file and checks its structure against the proposal.