Is your llms.txt useful?

llms.txt is a proposed Markdown file at the root of a site, or of a documentation section, that tells a language model where the useful content is. The proposal asks for one H1 with the project name, an optional blockquote summary, and H2 sections holding lists of named links with short notes. It is a community proposal, not an IETF or W3C standard, and no crawler is obliged to read it.

Six things. Whether the applicable file answers as text, whether the page points to it with rel=describedby, whether the first content line is one H1 title, whether a blockquote summary follows it, whether the H2 sections hold named Markdown links, and whether the first 20 unique linked resources answer as agent-friendly text. Only the file and the H1 are required. The other four are recommendations, so a missing one warns and does not fail.

A page can name its file with a Link header or an HTML link element carrying rel=describedby, and the checker reads that first. Without one, it walks from the page's own directory up to the site root and requests llms.txt at each step, so a docs section can have its own file while the root keeps another. If you enter the file address itself, it is read directly. The result names the route that found the file and every address it tried.

Each row carries the exact line to add. A missing file gets a starter with the H1 and the summary. A file that returns HTML usually means a catch-all route answered instead of a text file, so serve plain Markdown at that address or return 404. Links that lead to HTML pages should point at Markdown or plain-text versions where you have them. IndieLLMs reads at most 512 KB of the file and requests only the first 20 links, and it says how many it skipped.

After a page check, IndieLLMs builds an editable starter draft from the page's title tag, its description tag, and the same-site links you tick. Anything it had to guess is marked as inferred. Copy or download the draft, then publish it as /llms.txt and check again.

Not in any announced way. Google has said it does not use the file, and no major AI provider has committed to reading it. A valid llms.txt grants no crawler access, blocks nothing, and guarantees no citation; robots.txt still decides who may crawl. IndieLLMs grades the file against the proposal so it is correct if a reader ever comes, and the result says exactly that in its caveat.

Yes. No account, no card, no premium plan, no upgrade nag, no hidden paywall. IndieLLMs is free, all of it.

There is no public lookup history. A network check sends the public input needed for that check to its named source. Local file tools process file contents in your browser. Analytics is currently off, so no visit or job data is sent to a tracker. Anything the tool saves for you stays in your browser.