AI crawlers read a plain text file called llms.txt to understand what your site is and which pages matter. Fill the fields, copy the file, upload it to your domain root. Nothing is stored, nothing is sent anywhere.
Upload it so it answers at yourdomain.com/llms.txt. It must be plain text, served at the root, and readable without login.
It is a plain text file at the root of a site that gives an AI system a short map: what the site covers, which pages are the important ones, and one line about each. Think of it as a sitemap written for a reader instead of a crawler. It is a proposal rather than an official standard, which is why it costs nothing to add and nothing to remove.
Assistant answers are assembled from pages the model can find and quote. A short, honest map makes it easier for a crawler to reach the pages that answer a question and skip the ones that do not. On a site with a few hundred URLs it changes little; on a small site with ten good pages and thirty thin ones, it does the sorting for you.
The file must answer at /llms.txt on your own domain. That is it. No DNS change, no script tag, no plugin. On a static host it is one file upload; on WordPress it is a file in the site root.
No. A sitemap lists URLs for search engines. llms.txt describes them for assistants. Keep both. The sitemap is for Google, the llms.txt is for the model that answers the question after someone asks it.
The free AI audit runs the same checks on your site: llms.txt, robots.txt access for AI crawlers, schema, and whether an assistant can quote your answer in the first hundred words. It takes 48 hours and it is free.
A plain text file that lists your key pages so AI crawlers can find them quickly.
At the root of your domain, as /llms.txt.
Not all of them yet. It costs nothing to publish and the ones that read it use it.
No. robots.txt sets permission, llms.txt offers a map.