llms.txt generator

What is llms.txt?

llms.txt is a proposed convention for handing a language model a clean, linkable map of your site: what you are, in one line, and the handful of pages worth reading first. It sits at the root of the domain, next to robots.txt, and it is plain text.

The generator above fetches the URL you give it, reads the real navigation and headings, and writes the file. It is the same code path the full audit uses, exposed on its own so you can use it without an account.

Does anything actually read it?

Adoption is not universal and we will not pretend otherwise. What is true is that it costs one static file, it cannot hurt you, and writing it forces you to answer the two questions the checks that definitely matter also ask: what is this site, and which pages are worth quoting.

If you only ever do one thing with the output, do this: read the first three lines and see whether they describe your business the way you would describe it out loud. If they do not, that is what a model currently thinks you are.

Where do I put the file?

Save it as llms.txt and serve it at the root of your domain, so that yourdomain.com/llms.txt returns it as text/plain. Any static host will do this without configuration. If your framework needs a route for it, that route should return the file contents unchanged — an HTML page that happens to contain the text does not count, and our audit will not score it.