Why it matters
llms.txt is a short Markdown file at the root of a site that tells AI assistants what the site is and which pages matter, in the format proposed at llmstxt.org: a title, a one-line summary and lists of links under headings. It's easy to get subtly wrong: relative links that break out of context, the same page listed twice, prose where links should be, or a huge file that should have been llms-full.txt.
This validator reads the file line by line against that format and shows the outline the way a parser sees it, every link with its problem, and a cleaned-up copy. Be clear-eyed about it: no AI company has promised to read llms.txt. It's cheap to add and harmless, and a tidy one is more likely to be useful if they do.
How it works
- InputYour llms.txt, pasted or fetched
- In your browserParsed against the llmstxt.org format
- OutputErrors, link checks and a cleaned file
Where it fits in a designer's workflow
- Plan
- Design
- Build
- Test
- Launch
By the numbers
Show the numbers as a table
| Link | Check |
|---|---|
| Dining tables | ok |
| Chairs | relative |
| Sideboards | ok |
| Dining tables | duplicate |
| Caring for oak | ok |
| Delivery | ok |
| Returns policy | ok |
| Our story | ok |
How to use it
- Paste your llms.txt, or enter your site and press Fetch.
- Fix the errors first (red), then the warnings.
- Check the links table for relative and duplicate links.
- Download the cleaned file and upload it to /llms.txt.
FAQ
Is llms.txt a standard?
It's a proposal from llmstxt.org, used by a growing number of sites and documentation tools. No search engine or AI company has committed to using it.
What's the difference from llms-full.txt?
llms.txt is a short index of links. llms-full.txt holds the full text of the important pages in one file. Build one with the llms-full.txt Builder.
Why are relative links a problem?
AI tools often read the file out of context, without knowing which site it came from, so /tables means nothing. Full URLs always work.
Does it change my robots.txt?
No. llms.txt doesn't control crawling. Use robots.txt for that.