Do you have a real llms.txt?
Enter a domain and see what an AI assistant actually gets when it asks for your llms.txt, plus a file generated from your own website to copy.
What this tool does
An llms.txt is a plain text file at the root of a website that tells AI assistants what the site is and which pages matter. This tool asks your domain for it and tells you which of these you have: a real guide, a file written by an SEO plugin, a list of bot permissions, an empty file, a soft 200 (your server answers 200 and returns your homepage, which looks fine in a browser and is not a file at all), a firewall blocking us, or nothing. Then it generates an llms.txt from your own website, with every unverified fact left as a marked placeholder, ready to copy.
Frequently asked questions
It is a plain text file published at the root of a website, at /llms.txt, that describes in plain language what the site is, what it offers and which pages matter. It exists so an AI assistant can read the important facts without crawling and guessing from your navigation.
A soft 200 is a server answering 200 OK at /llms.txt and returning your homepage HTML instead of a text file. In a browser it looks like the page loaded, so nobody notices, and to a machine there is no file at all. It is the most common surprise we find, and the fix belongs to your hosting or your router rules.
It is not a ranking factor and no assistant promises to read it. What it does is remove ambiguity: the facts you want quoted are written once, in your own words, in a file that is trivial to parse. It is cheap to publish and it costs nothing to be wrong about.
At the root of your domain, so it opens at yourdomain.com/llms.txt, served as plain text. If you publish a second language, give it its own path such as /llms-es.txt rather than overwriting the canonical one, because only one file can live at /llms.txt.
Because we only fill in what we could verify by reading your website. Anything we could not check is left as a marked placeholder instead of a plausible guess. A generated file full of invented facts would be worse than nothing, since the whole point is to be the source an assistant trusts.
No, and mixing them is a common mistake. robots.txt says who may read your site. llms.txt says what your site is. If your llms.txt is full of User-agent and Disallow lines, those rules belong in robots.txt and your llms.txt is not describing anything.