Skip to content
Vouch

llms.txt validator

Check a live or pasted llms.txt against the llmstxt.org proposal and get a pass or fail list, with a fix for every issue. Free, no sign-up.

Check a site's llms.txt

Paste your llms.txt

Not live yet? Paste the file here. The check runs in your browser, and nothing you paste is sent to us. For a file that is live, use the site check above: it also checks how the file is served.

Up to 512,000 characters.

What the llms.txt validator checks

The checks follow the llmstxt.org proposal. Only the H1 is required; the rest are the proposal's own conventions, so the validator calls them problems when a tool reading by the proposal would be misled and suggestions when the file is merely thin.

  • An H1 with the site's name is the first line, and there is only one.
  • A blockquote under it summarises the site (a suggestion if missing).
  • No heading is below H2, and each H2 section holds a list of links.
  • Every link has a full https:// address, not a relative path.
  • An ## Optional section, if there is one, comes last.
  • For a live file: status 200, Markdown or plain text rather than an HTML page, and under 512 KB.

A broken file and a valid one

This file has text before the H1, a list item that is not a link and a relative address:

Welcome to Example
# Example
## Docs
- Getting started
- [Pricing](/pricing)

This one passes every check:

# Example

> Example is a reporting tool for subscription businesses.

## Docs

- [Getting started](https://example.com/docs/start): Connect billing
- [Pricing](https://example.com/pricing): Plans and prices

After it passes

No file yet? Build one with the generator. Then check the bigger question, whether the AI crawlers may read your site at all. For what assistants say once they can, see Vouch's plans, or read how llms.txt fits into GEO first. All the free GEO tools are listed together.

Common questions

How do I validate my llms.txt file?

Type your site's address and press Check the file. The validator reads https://your-site/llms.txt the way a tool would and checks it against the llmstxt.org proposal: an H1 naming the site first and only one, a blockquote summary, no heading below H2, H2 sections of Markdown links with full http or https addresses, and an Optional section last. It also checks that the address serves Markdown, not an HTML page. If the file is not live yet, paste it instead.

What are the most common llms.txt errors?

The address returns an HTML page (a catch-all route or a themed not-found page answering with status 200), the file has no H1 or text above it, list items under an H2 are plain text instead of links, links are relative paths, or headings go below H2. Each one is reported with a fix.

How to write an llms.txt file, with examples

Is llms.txt an official standard?

No. It is a 2024 proposal by Jeremy Howard, open for community input, and the validator checks against that proposal and nothing stricter. Google says its Search ignores the file. Coding agents and some other tools do read it, so treat it as a cheap, optional map.

llms.txt: what it is and whether it helps

Does a valid llms.txt get me cited by AI assistants?

Not by itself. A well-formed file is a tidy map; it does not make any assistant read it or name you. Whether assistants can read your pages at all is set in robots.txt, and what they then say about you is what Vouch measures. Start by checking that the main AI crawlers are allowed.

AI crawler checker

What does the validator store?

A site's result is kept in memory for an hour, so repeat checks and shared result links do not hit the site again. Nothing about the site is saved to a database. What you paste stays in your browser and is never sent to us. The address of the person checking is only counted, hashed, for the rate limit. The validator fetches only public web addresses and refuses private or internal ones.