llms.txt validator

The order and shape the llmstxt.org proposal describes: an H1 with the site or project name first (the only required part), an optional blockquote summary, optional detail without headings, then H2 sections whose list items each start with a Markdown link [name](url), optionally followed by a colon and notes. It also flags HTML served instead of Markdown, list items without links, empty sections and unclosed code blocks.

The usual place is the root, /llms.txt. The current proposal (v2) also allows a file at any path, such as /docs/llms.txt, which covers the pages under that path; when more than one file applies, agents should use the most specific one. Enter the full path to check a file that lives in a subfolder.

Your server returned HTML at that address, usually a 404 page or the homepage from a catch-all route. The file has to be served as plain text or Markdown. Upload it as a static file, or check the route or plugin that is meant to generate it.

No. Only a missing H1, an H1 that is not first, list items without a link, and an empty or HTML response break the format. Warnings, such as a missing summary or relative links, are suggestions that make the file easier for agents to use.

Pasted text is checked in your browser. If you check by URL, RankRoot's server fetches only the llms.txt or llms-full.txt file at that address and returns it to your browser for checking.