The short answer
llms.txt was proposed in September 2024 by Jeremy Howard of Answer.AI as a plain list of a site's most useful pages, written in Markdown so a model can read it cheaply. It sits alongside robots.txt and sitemap.xml but does a different job: robots.txt says what crawlers may fetch, sitemap.xml lists every page for search engines, and llms.txt is a curated reading list for language models.
What the file looks like
# Your Business > One sentence on what you do and for whom. ## Key pages - [Pricing](https://yourbusiness.com/pricing) - [Services](https://yourbusiness.com/services) - [Contact](https://yourbusiness.com/contact)
Does it help you get cited?
There is no public evidence that it does. OpenAI, Google, Anthropic and Perplexity have not said their crawlers read llms.txt, and the AI visibility research we rely on found no measured effect. In the sites our checker has read so far, 1 of 2 had an llms.txt at all.
Compare that with what does have evidence. In a controlled study of generative engines (Aggarwal et al., KDD 2024), adding quotations, statistics and cited sources to a page raised its visibility in answers by roughly 30–41%, while keyword stuffing lowered it by about 10%.
Should you add one?
- Yes, if it takes you ten minutes: it is harmless and may matter later. Our free llms.txt generator drafts one from your own pages.
- No, if it competes with real work. Ranking for the searches assistants run and getting listed on the sources they cite move answers today.
- Never block AI search crawlers in robots.txt while adding llms.txt: a blocked crawler cannot read either.
What to do instead
Find the questions buyers ask, see which searches the AI runs to answer them, and check whether you rank for those searches. That is what our free AI visibility check shows in about two minutes.
Sources: llmstxt.org (the proposal, September 2024); P. Aggarwal et al., "GEO: Generative Engine Optimization", KDD 2024, arXiv:2311.09735. Counts from Sightings's own site checks, October 2026.