No Signup Unlimited Usage Real Data 100% Free Privacy First Fast Results Export Reports Mobile Friendly

llms.txt Validator

Validate a live /llms.txt or pasted text against the llmstxt.org format: HTTP status, Content-Type, UTF-8 and size, the H1 and summary, list-item syntax with line numbers, every link (HEAD/GET), llms-full.txt and robots.txt conflicts. Then edit it in the generator.

Fetching live data…

Quick answer

The llms.txt Validator checks whether a site's llms.txt file is reachable, correctly structured and full of working links. Enter a URL or paste the file's text, and the tool tests it against the format described at llmstxt.org, lists every error and warning with its line number and a suggested fix, and checks each link it contains.

How it works

  1. Choose a mode: enter a site URL to fetch the live /llms.txt, or paste the text to validate a draft before you upload it.
  2. In URL mode the tool checks the HTTP status, Content-Type, encoding and size, and looks for /llms-full.txt as well.
  3. It parses the Markdown line by line: headings, the summary blockquote, sections and list items.
  4. It tests up to 50 links, five at a time, compares the file with your robots.txt, and returns a score, an issue list, a structure tree and a link table.

What it checks

  • Access — /llms.txt must answer with status 200. A Content-Type other than text/plain or text/markdown is a warning.
  • Encoding and size — the file must be valid UTF-8. Over 100 KB is a warning; over 1 MB is an error.
  • llms-full.txt — whether the optional full-content companion file exists.
  • H1 title — the first meaningful line must be exactly one H1. This is required, and a missing H1 is an error.
  • Summary — a > blockquote summary should follow the title. It is recommended rather than required.
  • Heading order — no H3 or deeper heading may appear before the first H2, and empty H2 sections are flagged as warnings.
  • List items — links under an H2 must look like - [name](url) or - [name](url): note. Malformed lines are reported with their line numbers.
  • Optional section — recognises ## Optional, the section the format reserves for links that can be skipped when context is short.
  • Links — each URL is requested with HEAD, falling back to GET when HEAD returns 405. A 4xx or 5xx response is an error; a relative URL is a warning.

Why it matters

An llms.txt file is a short, curated map of a site written for language models. The llms.txt proposal keeps the format deliberately simple, which makes small mistakes easy to miss: a second H1, a stray H3 above the first section, a list item without brackets, or a link to a page that was moved last year. A file that a parser cannot read cleanly, or that points to broken pages, does little of the job it was written for.

The validator also looks one step further. If your robots.txt fully blocks GPTBot, ClaudeBot or PerplexityBot, it warns you: the llms.txt then points AI services to pages those bots may not be allowed to crawl. Check each bot's access in detail with the AI Crawler Access Checker.

How to use the results

The score has three parts: required rules are worth 40 points, recommended structure 30, and the remaining 30 are multiplied by the share of links that work. It is our own measure of how closely the file follows the published format, not a rule set by any AI company, and no AI service has to read or follow an llms.txt file.

  • Fix errors first — a missing H1, an unreachable file or broken links. Each issue shows its line number and a suggested change.
  • Then warnings — add the summary blockquote, convert relative URLs to absolute ones, and fill or remove empty sections.
  • Edit in Generator — this button opens the llms.txt Generator (AI / GEO) pre-filled with the parsed title, summary and links, so you can correct the file and export a clean copy.
  • Check the linked pages — most AI crawlers do not run JavaScript, so test your key URLs with the JavaScript Rendering Checker to confirm their text is in the raw HTML.

Frequently asked questions

Do I need an llms.txt file?

It is optional. llms.txt is a proposed format, not an official standard, and AI services decide for themselves whether to read it. If you publish one, though, it should be valid and its links should work, because a broken file sends models to the wrong places or nowhere at all.

Can I validate a file before uploading it?

Yes. Switch to the paste mode and drop in the text. The structure and link checks run on the pasted content; the access checks for status, Content-Type and size apply only when the tool fetches a live file from a URL.

Why is a relative URL only a warning?

A relative link such as /docs can be resolved against your domain, but a model reading the file on its own may not know which domain to use. Absolute URLs remove that ambiguity, so the validator recommends them without treating relative links as broken.

What does the Optional section do?

The format gives ## Optional a special meaning: links listed there are secondary and can be skipped when a shorter context is needed. The validator recognises the section and checks its list items and links the same way as every other section.

Why does it compare llms.txt with robots.txt?

The two files can contradict each other. If robots.txt fully blocks GPTBot, ClaudeBot or PerplexityBot, your llms.txt recommends pages those bots may not crawl. The warning helps you decide which signal you actually want to send.

Is the llms.txt Validator free?

Yes, it is free to use in both modes. Paste a draft as often as you like, or enter a URL to check the live file, its links and its companion files. For a broader AI-readiness view, run the AI Visibility Checker.

Learn & Resources

Guides, tips and tutorials hand-picked for this tool — go from running a check to actually fixing it.

What Is SEO? A Complete Guide for Beginners — illustrated guideOn-Page SEO
Jul 10, 2026 · 15 min read

What Is SEO? A Complete Guide for Beginners

SEO is how you make a website rank higher in search engines so it earns free, organic traffic. This complete beginner's guide explains what SEO is, how search engines work, the four pillars (keyword research, on-page, technical and off-page), GEO for AI search, and a simple step-by-step roadmap you can start today.

Read article →