Check your llms.txt actually parses.
An llms.txt is a curated, machine-readable index of your site: an H1 with your name, a one-line summary, then sections of absolute links with short notes. Get the structure wrong and a tool reading it gets nothing useful. This checks yours against the published format, missing or duplicate H1, no summary, relative URLs, bare links, empty sections, a misplaced ## Optional block, and shows the first thing an agent would read. Runs entirely in your browser, no key, nothing sent to us.
A clean index, or noise.
Open yourdomain.com/llms.txt, select all, copy, and paste it below. Don’t have one yet? The Schema & llms.txt Generator will write one for you, then bring it back here to check it.
Parsed entirely in your browser, nothing is uploaded, logged, or stored. This checks structure against the published llms.txt format; the spec is young and engine support varies, so a valid file is a clean signal, not a guarantee of citation. Want it built and maintained? Talk to us.
Questions, answered.
A plain-text, curated index of your site at yourdomain.com/llms.txt: an H1 with your name, a one-line summary, then sections of absolute links with short notes. It is meant to give a language model a clean map of your best pages instead of making it crawl and guess.
There is no evidence it does on its own, and we would not claim otherwise. Support is partial and the spec is young. It is cheap to maintain and it removes ambiguity about what your key pages are, treat it as tidy plumbing, not a ranking lever.
The generator WRITES a starter file from a short form. This validator CHECKS a file that already exists, structure, link format, sections, and the mistakes that make it unparseable. Generate once, validate after every change.
A missing or duplicated H1, a missing summary, relative URLs (the file is read out of context so every link must be absolute), bare URLs that are not markdown links, links sitting outside any section, empty sections, duplicates, and a '## Optional' block that is not last.
No. It is a curated index, not a sitemap. Keep it to the pages you would want an assistant to read first, and put long-form content in llms-full.txt if you maintain one.