Glossary

What is llms.txt?

llms.txt is a plain-text file at the root of a site that gives language models a short, readable map of who runs the site and which pages matter.

It is a proposed convention, published at llmstxt.org, not a standard that search engines have agreed to. The file is written in Markdown: a title, a one-line summary, then sections of links with a short description after each one. Some sites also publish a longer companion file, often called llms-full.txt, with more detail.

Nobody can tell you which AI systems read the file, and we do not claim any do. What it does is cheap and low-risk: it puts a clean description of the site where a system that looks for it can find it, without menus, scripts or cookie banners in the way.

The main risk is drift. A hand-written llms.txt goes out of date the first time a page is renamed or removed, and then it describes a site that no longer exists. That is worse than having no file. The fix is to generate it from the same list of routes that builds the sitemap, so the two can never disagree.

An example from our own work

On audacian.eu both files are generated at build time. audacian.eu/llms.txt opens with one sentence about who we are, then lists our work grouped by service (Web, AI, Video, SEO, Extensions), our insights articles and the translated URLs of the main pages. audacian.eu/llms-full.txt lists every indexable URL with the exact title and meta description that page ships, and skips any page marked noindex. Both come from the same route list as the sitemap, so a removed page disappears from all three at once. Our robots.txt points at both files in a comment. Videojam, our own video platform, generates its llms.txt from the same indexation gate as its sitemap, so pages kept out of the index are kept out of llms.txt too.

Tell us what you're launching.

Book a 20-minute intro call, or email a few lines about what you're launching. We'll reply with a rough plan and a price range, whether or not we end up working together.