llms.txt Checker: Validate Any Site's llms.txt File

What the llms.txt checker tests

How to read your score

The score starts at 100 and each problem takes points off. A missing title costs the most. A missing summary, no link sections or an unusable URL cost 10 to 20 points, and smaller format slips cost 2 to 5. Notes cost nothing: they are suggestions, such as adding an Optional section. A missing file, an HTML page or a server error scores 0.

• 90 to 100, good: the file follows the format closely.
• 50 to 89, needs work: tools can read it, but some lines will be skipped or misread.
• Below 50, poor: the file is missing its core parts.

The score is our own measure of how closely a file follows the proposal. No AI company publishes a scoring rule for llms.txt, so treat it as a checklist, not a ranking signal.

Common llms.txt errors and how to fix them

• The address returns a web page. Your framework's catch-all route or 404 page answers /llms.txt, sometimes with status 200. Put a real file in the folder your site serves static files from (public/ in Next.js) so the address returns text.
• Request refused (403 or 429). Bot protection or a firewall rule challenges requests that do not come from a browser, so tools never see the file. Allow /llms.txt: it is a public file meant for automated readers.
• Wrong content type. A file labelled text/html or application/octet-stream may show as one long paragraph or download instead of opening. Serve it as text/plain; charset=utf-8 or text/markdown; charset=utf-8. Most hosts do this automatically for a .txt file.
• Plain text inside sections. Instructions and bullet points without links belong above the first H2. Inside a section, every item should start with [name](url).
• Relative links. /pricing only works for a reader that knows where the file lives. Write https://example.com/pricing.
• Notes after a dash. The format puts a colon between a link and its note: - [Pricing](https://example.com/pricing): Plans and limits.
• A copy of the sitemap. Hundreds of links defeat the purpose. Keep the pages that answer real questions and move the rest under ## Optional.

llms.txt vs robots.txt vs sitemap.xml

• robots.txt tells crawlers which URLs they may fetch. It is the file that allows or blocks AI bots such as GPTBot or ClaudeBot; check yours with the AI crawler checker.
• sitemap.xml lists the URLs you want search engines to know about, every one of them.
• llms.txt is a curated, readable overview of your most useful pages for language models. It does not allow or block anything.

They work side by side, and llms.txt does not replace either one.

Who reads llms.txt today

Be clear-eyed about it. Google's guide to its generative AI features says it is fine to keep an llms.txt file, but Google Search ignores it, so it neither helps nor hurts your rankings. We have not found documentation from other AI search products saying they use it to choose sources.

Other readers do exist. AI companies publish llms.txt files for their own developer docs (OpenAI and Anthropic do), and so do companies such as Stripe, Vercel and GitHub. Chrome's Lighthouse has an experimental Agentic Browsing category with an llms.txt audit: it flags a server error on the file and marks a missing file as not applicable. Coding assistants and AI agents can read the file when someone points them at your site.

So llms.txt is worth the hour it takes, and no substitute for clear pages and mentions on sites people trust. The llms.txt guide covers the rest.