Definition
llms.txt is a proposed plain-text file at a site's root (like robots.txt) that gives large language models a curated, markdown-formatted map of the site's most important content. Proposed by Jeremy Howard in 2024 (llmstxt.org), it lets sites present clean, token-efficient context to AI systems instead of forcing them to parse full HTML. The format is deliberately simple: an H1 with the project name, a short summary, then curated markdown links grouped by section; a companion llms-full.txt can carry expanded content. Adoption is voluntary and engine support varies, so treat llms.txt as a complement to — not a replacement for — robots.txt, sitemaps and schema: engines that ignore it lose nothing, and engines that read it get your best pages in clean form. Publishing one is a low-cost way to control the narrative AI models see first.