FREE · LIVE TEST · NO SIGN-UP
FREE AI Visibility Checker
Live test from our servers, no sign-up, about 15 seconds. We check your site for 33 AI crawlers, from robots.txt to your CDN and server, and show you exactly what blocks them, and how to fix it.
HOW IT WORKS
We check your site for 33 different AI crawlers like ChatGPT, Claude & Perplexity
Then we compare their answers with what a normal visitor sees. If ChatGPT, Claude, Perplexity or ANY other AI crawler is turned away while a person walks in, you’ll see who was blocked, where, and how to fix it.
- 01Visitor checkWe load your site like a normal browser to see what people get.
- 02Read robots.txtWe check what your robots.txt file tells each AI it may see.
- 03Check for 33 AIsChatGPT, Claude, Perplexity, Google and more, one by one.
- 04Find the blockerIf an AI is turned away, we pinpoint exactly what stopped it.
- 05Get the fixClear steps for you or your developer, or a prompt for ChatGPT.
What your score means
- 0–29 · HiddenHidden from AI search
- 30–59 · Largely hiddenA CDN or WAF rule blocks AI search
- 60–79 · Partly hiddenGeneric bot detection or a shared rule
- 80–94 · Mostly visibleOne or two fetchers blocked
- 95–100 · Fully visibleNothing blocked
WHY MOST CHECKERS GET THIS WRONG
robots.txt says allowed. But the crawler gets a 403.
User-agent: ChatGPT-Bot Allow: /✓A robots.txt checker says you’re fine
HTTP/2 403 server: cloudflare title: Attention Required! | Cloudflare✕Blocked at the edge, invisible to ChatGPT
Since Cloudflare, Vercel, Akamai and the other edge networks started blocking AI crawlers with rules that never appear in robots.txt, reading the file is not enough. We check your live site for 33 AI crawlers & bots.
Is your website blocking ChatGPT, Claude or Perplexity?
robots.txt can’t tell you on its own. Cloudflare and other CDNs block AI crawlers with firewall rules that never appear in the file, so we check your live site for each crawler below, not just the file. Each AI company runs separate bots for search, for opening a page when someone asks about it, and for training, so blocking one says nothing about the others.
Which bots do ChatGPT, Claude and Perplexity use?
ChatGPT
- Finds and cites youOAI-SearchBot
- Opens your page when askedChatGPT-User
- TrainingGPTBot
Claude
- Finds and cites youClaude-SearchBot
- Opens your page when askedClaude-User
- TrainingClaudeBot
Perplexity
- Finds and cites youPerplexityBot
- Opens your page when askedPerplexity-User
- TrainingNone
Google AI Overviews, AI Mode & Gemini
- Finds and cites youGooglebot
- Opens your page when askedGoogle-AgentGoogle-GeminiNotebook
- TrainingGoogle-Extended
Meta AI
- Finds and cites youmeta-webindexer
- Opens your page when askedmeta-externalfetcher
- Trainingmeta-externalagent
Mistral
- Finds and cites youMistralAI-Index
- Opens your page when askedMistralAI-User
- TrainingMistralAI-Training
Alexa
- Finds and cites youAmzn-SearchBot
- Opens your page when askedAmzn-User
- TrainingAmazonbot
Siri & Apple Intelligence
- Finds and cites youApplebot
- TrainingApplebot-Extended
Microsoft Copilot
- Finds and cites youbingbot
DuckDuckGo AI answers
- Finds and cites youDuckAssistBot
You.com
- Finds and cites youYouBot
AI search crawlers
scoredDecide whether ChatGPT search, Claude, Perplexity, AI Overviews, Copilot and Siri can find and cite you.
AI assistant user agents
scoredThe user-triggered fetchers that open your page when someone asks an assistant about you.
AI training bots
policy · not scoredReported, never scored. Blocking them is your choice about model training.
ANATOMY OF A RESULT
What you get back in 15 seconds
A score, a verdict for every crawler, the exact layer doing the blocking and a fix plan written for that layer. Every verdict comes with its evidence.
- 01A score out of 100AI search and assistant access in one number, with a share link that keeps for 30 days.
- 02The layer that’s blocking youCloudflare, Vercel, Akamai, AWS WAF, your server or robots.txt, named so you open the right dashboard.
- 03Evidence you can reproduceThe response or robots.txt line behind every verdict. Rerun it yourself with one curl command.
- 04A fix plan for your setupSteps for that layer, a note for your developer and a ready prompt for ChatGPT or Claude.
- 05Monitoring, if you want itWe keep re-checking and email you the moment a crawler is blocked or let back in.
- 06A clear next stepReachable isn’t the same as recommended. Every result ends with what gets brands cited.
FAQ
What does the AI Visibility Checker test?
It checks your site for 33 AI crawlers. From our own servers it reads your robots.txt and fetches your homepage, plus one deeper page when robots.txt names a path or you entered one, first as a real browser and as our own honestly named bot, then with the exact user agents of the AI search, assistant and main training crawlers (OAI-SearchBot, ClaudeBot, PerplexityBot, GPTBot, Googlebot and the rest), and records how each crawler is answered: allowed, blocked by robots.txt, blocked at the edge by a CDN or WAF rule, challenged by bot detection, dropped, or paid access. It then names the layer that produced the block and gives you the fix for that layer.
Is my website blocking ChatGPT?
Check three bots, not one. OAI-SearchBot decides whether ChatGPT search can find and cite your pages, ChatGPT-User opens a page live when someone asks ChatGPT about it, and GPTBot only collects training data. Blocking GPTBot keeps your content out of OpenAI's model training but not out of ChatGPT's answers; blocking OAI-SearchBot or ChatGPT-User does. The checker shows what each of the three actually gets from your site, not just what robots.txt says.
Is my website blocking Claude?
Anthropic uses the same split. Claude-SearchBot indexes pages for Claude's search results, Claude-User fetches a page when someone asks Claude to read it, and ClaudeBot collects training data. If Claude-SearchBot or Claude-User is refused, Claude cannot read or cite your pages; blocking ClaudeBot only opts you out of training. The checker tests all three.
Can my website block AI bots even if robots.txt allows them?
Yes, and a robots.txt checker cannot see it. Since 2025 Cloudflare, Vercel, Akamai and other edge providers have blocked AI crawlers with firewall rules that never appear in robots.txt, and in September 2026 Cloudflare moved its default blocks from robots.txt lines to edge rules. A robots.txt-only checker reports "allowed" while the crawler is actually served a 403. We test the live response and compare it with what a browser gets.
Should I block GPTBot and other AI training bots?
That is a business decision, which is why the checker reports training crawlers but never scores them. Blocking GPTBot, ClaudeBot, CCBot or Google-Extended keeps your content out of model training. It does not remove you from ChatGPT, Claude or Google AI Overviews answers, which use different crawlers; blocking those search and assistant crawlers does.
What is the AI visibility score?
A 0-100 score computed from the AI search crawlers and the user-triggered assistant fetchers only, weighted by how much each one matters for being found and cited. Training crawlers (GPTBot, ClaudeBot, CCBot and so on) are shown in a separate policy panel and never lower the score: blocking training is a legitimate choice, blocking search is self-harm.
Is this an AI search visibility checker or an LLM visibility checker?
It is the access half of both. It tells you whether ChatGPT search, Claude, Perplexity, Google AI Overviews and AI Mode, Copilot and the AI assistants can fetch your pages at all, and what stops them. Getting recommended once you are reachable is a separate job, which is what the LinksThatRank AI Visibility packages do.
How accurate are the results?
Exact for user-agent-keyed rules and for robots.txt, which is how Cloudflare AI Crawl Control, the Vercel AI Bots ruleset, .htaccess rules and most WAF user-agent rules work. Our requests are unverified (they come from our servers, not from OpenAI's or Anthropic's IP ranges), so for rules keyed on verified identity the result is an upper bound. When a site treats every unknown client the same we say "challenged" or "could not test", never "blocked", and every verdict carries its evidence.
Is the checker free and do I need to sign up?
It is free, there is no sign-up and results are never gated. An email address is only asked for if you switch on monitoring.
Will you crawl my site?
No. Your robots.txt, your homepage and at most one deeper page: about 40 small requests per check, at most four at a time and spaced a few hundred milliseconds apart. Repeat checks of the same site are limited, and everyone else sees the stored result. If you would rather not be tested at all, the about page explains how to add your domain to the remove list.
Can I see a result again later?
Yes. Every result has its own link (copy it from the result card) with a preview that shows the score and the headline finding. Results are kept for 30 days.