llms.txt Checker
Check your llms.txt against the spec and test its links, then fix what fails. Free, no signup.
What the llms.txt checker tests
Enter a domain and the checker requests /llms.txt from its root, then runs five checks.
The response. The checker reports the status code, the content type and the file size. A 404 reads as "no file found". A 200 that returns an HTML page is flagged, because that usually means your app answered with a page instead of the file.
The structure. The llmstxt.org specification lists the parts in order: an H1 with the name of the site, an optional blockquote summary, optional free text, and optional H2 sections that hold lists of links. The H1 is the only required part. The checker follows that rule and reports a missing blockquote as information rather than an error.
The link format. Each item in a file list needs a Markdown hyperlink, [name](url), with optional notes after a colon. The checker flags bare URLs and links that sit outside an H2 section.
The links. The checker requests each listed URL, six at a time, and reports its status, any redirect and any failure, so a renamed page shows up by name. It tests up to 250 links. A file with 250 links or fewer is checked in full. Addresses that are not web links, such as mailto: addresses, are listed but not requested. On a larger file the checker tests 250 links spread evenly through the file and says so in the result, for example "Checked 250 of 821 links, spread evenly through the file".
The companion file. The checker reports whether /llms-full.txt exists. The spec text does not define that file. Documentation platforms often generate it, and the result reads "present" or "absent" with no score attached. It warns in two cases: the address returns a web page instead of a text file, or the file is very large.
The spec allows an llms.txt at the site root or at any path, covering the pages under that path. This checker tests the root file. It also leaves out the spec's proposal for clean Markdown copies of each page, and so a pass here says nothing about those.
Does anyone read llms.txt?
Check this before you spend a day on the file. llms.txt is a proposal published by Jeremy Howard on September 3, 2024. It remains a proposal, and no major search engine has said it reads the file to build answers. Our guide to who reads llms.txt keeps the sources in one place.
Log data agrees. Ahrefs studied 137,210 domains in May 2026. About 28 percent published an llms.txt, and 97 percent of those files received no requests that month. Ahrefs called the file "largely decoration" for AI search visibility. Among named AI tools, GPTBot fetched it most and Claude Code came second, and Ahrefs notes that "fetched" does not mean "read".
That leaves a narrow case. A file may help coding agents and tools that read documentation, and it costs an afternoon by hand. Publish one for that reason if you run developer docs. We promise nothing about AI citations from it.
How to read the result
Some findings are information only, with no effect on the result: a missing summary, relative links, links outside an H2 section, and redirects to another page on your site. The checker reads the first 4 MB of a file.
Common failures and how to fix them
A 200 response with an HTML page
Single-page apps often answer every unknown path with index.html. Serve llms.txt as a static file from your public folder or add a route for it.
No H1
Make the first line # Your Site Name. It is the one element the spec requires.
Bare URLs
Write each entry as - [Page title](https://example.com/page): one line on what it covers.
Links outside an H2 section
The spec puts file lists under H2 headings. Group links under headings such as ## Docs and ## Guides.
Dead links
A removed or renamed page returns a 404 or lands on your homepage by redirect. Update the link or drop it. Files written once and left alone tend to collect dead links.
403 or 429 on links that load in a browser
Bot protection can block the checker's request. Open the URL yourself. If it loads, allow the request or ignore that result.
Relative URLs
The spec text we read does not address them. Use absolute URLs so any agent can follow the link without guessing the base.
A long file
The spec's convention for skippable material is an ## Optional section, which tells an agent it can drop those links when it needs a shorter context. A Common Crawl analysis of 584,107 files found median sizes from 148 to 8,772 tokens depending on the generator. Our advice is to list the pages you would send a new reader to, then stop.
After the fixes, run the checker again. If you have no file yet, draft one with the llms.txt generator.
Tenpoint's angle
llms.txt tells an agent where your pages are. It says nothing about what AI engines say about you once they find them. Tenpoint Labs measures that: we ask ChatGPT, Perplexity, Gemini and Claude the questions your buyers ask and record who gets named, in what position and with what framing. The free AI visibility audit gives you a first read.
More free checks: check which AI crawlers can reach your site and validate your schema markup.
Key facts
A plain summary of the checker, written for people and for the systems that summarize pages.
- File tested
- /llms.txt at the site root, plus whether /llms-full.txt exists
- Checks
- Response, structure against the llmstxt.org spec, link format, links, companion file
- Links tested
- Up to 250, six at a time; an even sample on larger files
- Labels
- Pass, warning, error, and no file found
- Price
- Free, no signup, no email
- Made by
- Tenpoint Labs
Questions about llms.txt
What is an llms.txt checker?
An llms.txt checker fetches the llms.txt file on a domain and tests it against the llmstxt.org specification. This one reports the response, the structure, the link format, the status of up to 250 listed links and whether an llms-full.txt exists. It is free and needs no signup.
What is llms.txt?
llms.txt is a Markdown file at a site's root that names the site and lists its most useful pages with short notes. Jeremy Howard proposed it on September 3, 2024. It remains a proposal.
Does llms.txt help my site get cited by ChatGPT, Perplexity or Google?
No evidence supports that. Google says Search does not use llms.txt files and that keeping one won't help or hurt your visibility. OpenAI, Anthropic and Perplexity publish the file for their own docs and have made no statement that their crawlers read it elsewhere. In Ahrefs' study of 137,210 domains, 97 percent of the files received no requests in May 2026.
Is the summary blockquote required?
No. The spec says the H1 is the only required section, and the blockquote summary is optional. Some checkers flag a missing summary as an error. This one reports it as information.
Is there an official llms.txt validator?
We found no validator published at llmstxt.org. The specification is the reference, and this checker follows it. The spec page was last modified August 10, 2026, so a file that passed last year may deserve a fresh run.
What is the difference between llms.txt and robots.txt?
robots.txt carries allow and disallow rules that crawlers choose to follow. llms.txt carries a name, a summary and a list of links, and the spec defines no rules for blocking anything. Use the AI crawler checker to test which AI bots can reach your site.
Do I need an llms-full.txt?
The spec text we read does not define one. Documentation platforms often generate it as a single file of full page text. The checker reports whether yours exists, does not score it, and warns if it is served as a web page or is very large.
