Free · no sign-up
llms.txt checker
Paste a domain. We fetch its /llms.txt, check it against the format, and follow the links inside it — because a file full of 404s is the one failure nothing else in your stack will ever tell you about.
Report
The llms.txt format
Six parts, in this order. Only the first is required — everything the checker reports is measured against this.
| Part | Syntax | Status | What it does |
|---|---|---|---|
| H1 | # Project name | Required | The only required part of the file. Names whose docs these are. |
| Summary | > One or two sentences | Optional | A blockquote carrying what a reader needs to make sense of the rest. |
| Detail | Any markdown except headings | Optional | Paragraphs and lists. No headings — the next heading starts a section. |
| Sections | ## Section name | Optional | H2 only. Each one holds a file list and nothing deeper. |
| Links | - [Name](https://url): notes | Per section | A markdown list. The link is required; the note after the colon is not. |
| Optional | ## Optional | Convention | A section by this name marks links an agent may skip for a shorter context. |
The full specification lives at llmstxt.org.
A starter file
Valid as it stands. Swap the URLs, put it at the root of your domain, serve it as text/plain.
# Your Project > One or two sentences on what this is and who it is for. A reader > uses this to decide whether the rest is worth fetching. Anything else worth knowing before the links: how the docs are organised, what is out of date, what to read first. ## Docs - [Quickstart](https://example.com/docs/quickstart): The 10-minute path - [API reference](https://example.com/docs/api): Every endpoint and field - [Guides](https://example.com/docs/guides): Task-shaped walkthroughs ## Optional - [Changelog](https://example.com/changelog): Release history - [Blog](https://example.com/blog): Background and announcements
Questions
What is llms.txt?+
A markdown file at the root of your site — /llms.txt — that tells a language model which of your pages are worth reading and what each one covers. It exists because an LLM landing on your docs has to guess what matters; the file is you answering that directly, in a format small enough to sit in a context window.
What does this llms.txt checker look at?+
It fetches your /llms.txt, parses it against the format at llmstxt.org, and reports what a reader would actually get: whether the H1 and summary are there, whether sections and links are shaped correctly, whether the file is small enough to be useful — and whether the links in it resolve, which is the part nothing else in your stack ever tests.
Does llms.txt help my SEO?+
Not directly. No search engine has said it uses the file as a ranking signal, and Google has said publicly that it does not. What it does is give assistants that fetch your site a curated route through it instead of a guess. Treat it as documentation for machines, not as a ranking tactic.
Where does the file go?+
At the root: https://example.com/llms.txt. Serve it as text/plain or text/markdown. The common failure is a catch-all route returning your app's HTML shell instead of the file — this checker flags that, because it looks fine in a browser.
What is llms-full.txt?+
The companion file. Where llms.txt is an index of links, llms-full.txt carries the expanded content inline so a reader can take everything in one fetch. It is optional, and this checker tells you whether you have one.
How many links should it have?+
As many as are genuinely worth reading, and no more — the file is supposed to fit in a context window alongside whatever else the agent is holding. Past a dozen or so links, add an "Optional" section and move the secondary ones there, which is the format's own way of saying what can be skipped.
The file is for machines. The work still isn't.
Sokosumi's AI coworkers turn a brief into a finished file: the launch copy, the social set, the landing page.
*No Credit Card required