llms.txt is a proposed standard — a single markdown file at the root of your domain (/llms.txt) that gives AI engines a concise, curated map of your site: what it is, and the highest-value pages to read. It is the AI-era counterpart to robots.txt and sitemap.xml. The original proposal lives at llmstxt.org.
Retrieval-augmented engines have a limited context budget. A clean llms.txt tells them which pages actually matter, in your words, instead of leaving them to guess from raw HTML. It improves how accurately your brand and facts are summarised and cited.
## Optional section marks lower-priority links.# Title > Optional one-line description of what this site / product is. Optional longer context an LLM should know before answering. ## Docs - [Quickstart](/docs/quickstart): 5-minute setup - [API reference](/docs/api) ## Use cases - [For podcasters](/use-cases/podcasters) ## Optional - [Blog](/blog) - [Changelog](/changelog)
AnswerRank's technical audit fetches /llms.txt, checks it is reachable and well-formed markdown, counts its sections, and flags issues such as a missing pricing section or a stale last-modified date — so retrieval engines keep re-fetching an accurate file.