KhaqanWhat is llms-full.txt?
llms-full.txt is a companion to llms.txt that holds the complete text of a site's key pages in one Markdown file, so an AI assistant can load everything it needs in a single request instead of following links page by page.
Where llms.txt is an index, llms-full.txt is the book. The convention grew out of the original llms.txt proposal, which suggested that sites also publish clean Markdown versions of their pages. Documentation teams found it simpler to concatenate those pages into one file at /llms-full.txt, and the name stuck. There is no formal specification for it beyond that: it is plain Markdown, usually with each page separated by a heading, served with a text content type.
It matters for a business because assistants and developer tools often work with a single pasted document. A prospect can drop the file into a chat and ask what you sell, where you operate and how to contact you, and the answers come from your own words rather than from a scraped fragment. For an ERP integrator, a consultancy or a services company, that is the difference between being cited and being guessed at.
As an example, a company with a services page, three case studies and a pricing FAQ could publish an llms-full.txt that contains all five, regenerated by the same build step that renders the HTML. Keep it small enough that a model with a modest context window can take it whole; anything larger is simply truncated.
The same caution applies as for llms.txt: no major AI company has confirmed reading it, so treat it as a low-cost courtesy rather than a ranking lever. The common mistake is generating it once and never again, so it drifts away from the live site and ends up contradicting the pages it is meant to summarise.
Related terms
- llms.txt: llms.txt is a plain Markdown file placed at a website's root (/llms.txt) that gives AI assistants a short summary of the site and a curated list of its most useful pages, so a language model can find the right content quickly.
- AI search visibility: AI search visibility is how often, and how accurately, a website is used and cited by AI-driven search products such as ChatGPT search, Perplexity, Google AI Overviews, Copilot and Claude when people ask questions the site could answer.
- GEO (generative engine optimisation): Generative engine optimisation (GEO) is the practice of making a website easy for generative AI systems such as ChatGPT, Perplexity, Claude and Gemini to crawl, understand and cite, so that the site appears as a source in AI-generated answers.