llms.txt
llms.txt is an emerging plain-text file (served at /llms.txt) that tells large language models which pages, products, and facts a site wants cited in AI answers. Proposed by Answer.ai in 2024.
llms.txt is to AI crawlers what robots.txt is to Googlebot — a simple text file at the site root that guides the crawler. Unlike robots.txt, llms.txt does not block anything; it signals which canonical pages and facts matter for LLM responses.
A typical llms.txt lists: a 1–2 paragraph description of the company in language an LLM can quote verbatim, canonical URLs for key pillar pages and product pages, category positioning (what category you belong to, and how you describe it), and optional comparison context for how you relate to competitors.
The standard is not yet formally adopted by OpenAI, Anthropic, or Google, but LLM crawlers appear to pull llms.txt files when they encounter them. Publishing one is low-effort and an early AEO signal. Sprites ships an llms.txt alongside robots.txt — see https://www.sprites.ai/llms.txt.