# llms-full.txt ## Quick answer llms-full.txt is a proposed companion file to llms.txt that holds a site's full content in plain markdown, rather than just an index of links, intended to give an AI system everything it needs in one fetch instead of having to crawl each page separately. ## Why it matters Even a well-organized llms.txt is just a table of contents; an AI tool still has to fetch and parse every linked page individually to get the actual content. llms-full.txt matters, where it's supported, because it removes that extra step, though no major AI engine currently documents reading either file, so its practical effect today is limited and shouldn't be mistaken for a reliable visibility lever. The file also has real maintenance cost: a long markdown export of a site has to be regenerated whenever pages change, or it quietly becomes an out-of-date copy of the site that contradicts the live pages. Teams that publish one should generate it automatically from the same source as the site rather than editing it by hand. ## How to measure it There isn't a reliable way to measure whether llms-full.txt specifically is being read, since no major AI engine documents support for it and the practical effect of llms.txt files in general is unproven. Treat it as inexpensive housekeeping rather than a channel with a measurable return, and spend verification effort on things that are demonstrably crawled instead. If a team does publish the file, the only checks within its control are internal ones: confirm the file loads at the expected address, that its content matches the live pages, and that it is refreshed whenever the site changes. For visibility itself, keep tracking what AI engines actually say about the brand, since that answer, not the presence of a file, is the outcome that matters. ## Example A team spends a meeting debating whether to build a full llms-full.txt export for their documentation site. Given that no major AI engine currently documents reading the file, the more useful outcome of that meeting is deciding to spend the time on clean HTML and schema markup instead, which crawlers demonstrably do use far more reliably. ## FAQ ### Is llms-full.txt the same as llms.txt? No. llms.txt is a short index of links; llms-full.txt is a proposed fuller version containing the actual page content in markdown, not just links to it, aimed at saving an AI tool a separate fetch per page. ### Do AI engines actually read llms-full.txt? No major AI engine currently documents support for it, and the same is true of the related llms.txt file, so treat it as low-cost housekeeping rather than a reliable channel. ### Is it worth building an llms-full.txt file? It's cheap once a team already has the content, but clean HTML, a working sitemap and structured data are what crawlers demonstrably use, and are worth prioritizing first. ### Could llms-full.txt become more useful later? Possibly, if a major AI engine starts documenting support for it; until that happens, there's no evidence it changes how an AI engine describes a brand today. ## Related terms - /glossary/ai-crawlers - /glossary/crawl-budget - /glossary/schema-markup ## Related - Schema markup and AEO: /blog/schema-markup-and-aeo - Pricing: /pricing