Tools for the llms.txt standard
Free, browser-based tools: the generator crawls your website and builds a ready-made llms.txt. The validator checks format and content with a point score. REST API for developers.
Available tools
AI Visibility Check
Enter a domain — in seconds we rate how well your website is optimized for ChatGPT, Claude, Perplexity & Gemini. Score 0–100 across 5 categories plus concrete recommendations.
Open tool → LiveAI Citation Check
Is your website actually cited as a source by ChatGPT? We ask the AI live with web search and show whether you are cited, only mentioned or not found at all — including the sources used.
Open tool → NewAI Crawler Check
Does your robots.txt accidentally block GPTBot, ClaudeBot or PerplexityBot? We check 20 AI crawlers with real Google semantics — including the inheritance trap most checkers miss.
Open tool → Kit · €2.99AI Signal Kit
The check shows what is missing — the kit delivers the finished solution: Schema.org, Open Graph, robots.txt AI access and llms.txt as copy-&-paste building blocks. Universal for any website system, with step-by-step instructions.
Open tool → Generatorllms.txt Generator
Enter a domain — we crawl up to 100 pages, extract titles, meta descriptions and JSON-LD, and deliver a complete llms.txt to copy or download.
Open tool → Manualllms.txt by hand
No site to crawl? Enter project name, description and links directly — a finished llms.txt per the standard. The detailed llms-full.txt is included for free.
Open tool → Generatorllms-full.txt Generator
Builds the detailed variant with the full body text of every crawled page — the file AI systems read for full-text indexing.
Open tool → Validatorllms.txt Validator
Checks your existing llms.txt for reachability, format, required fields and size. Rating of 0–100 points and an A–F grade with detailed notes.
Open tool → APIREST API for developers
Validate domains programmatically. JSON responses, 60 requests per hour per IP, no API key required. Health endpoint included.
Documentation →About the tools
All in the browser, all free. The generator tools crawl your website live, parse the HTML content and deliver a Markdown file following the official llms.txt specification. No data is stored — processing runs fresh for every request.
Respects robots.txt. By default the crawlers check the target domain's robots.txt and skip blocked paths. If you crawl your own domain, you can disable this option.
Validator score. Nine individual checks produce a score of 0–100 and a grade (A–F): reachability, content type, Markdown title, description, sections, links, file size, UTF-8 and the reference to llms-full.txt.
Frequently asked questions
Are the tools really free?
Yes. The generator and validator have no limit other than the page cap per crawl. The REST API allows 60 requests per hour per IP address — enough for most uses.
Is my data stored?
No. The crawler tools store nothing permanently. The REST API only keeps a rate-limit counter per IP, no content or checked domains. Privacy: see the privacy policy.
How exactly does the generator work?
We start at the root path of the given domain and read the title, meta description, JSON-LD and body text. With deep scan enabled we follow internal links up to the maximum page count. From this data we build the Markdown structure (title, blockquote, sections with link lists).
Do I need both files — llms.txt and llms-full.txt?
llms.txt is required — a compact content overview. llms-full.txt is optional but recommended: the detailed full-text variant for better AI indexing. The validator awards bonus points when both are present.
What does the A–F grade mean?
A (90–100): exemplary · B (75–89): minor improvements possible · C (55–74): several recommendations · D (35–54): essential elements missing · F (0–34): not usable.