Is ChatGPT allowed to read your website?
Your customers ask AI assistants about you before they ever visit. If OpenAI's and Anthropic's crawlers are blocked (in robots.txt or at your CDN), your site can't shape what those assistants say. Check it in one click.
We read your public robots.txt and probe your homepage with AI bot user-agents. Nothing is stored.
Not all AI bots are the same
Blocking one type protects your content; blocking another makes you invisible. Here's the difference the checker uses.
Fetches your page in real time when someone asks the assistant about you. Block this and you vanish from the answer. Keep it open.
Indexes your site so it can be cited in AI search results (ChatGPT Search, Perplexity, etc.). Keep it open to stay discoverable.
Collects pages to train future models. Blocking it is a legitimate content/IP choice. Just know it does not remove you from live answers, and past training already happened.
Common questions
Why does it matter if AI crawlers can read my site?
Customers increasingly ask ChatGPT, Claude, and Perplexity “is this business any good?” or “best <category> near me?” instead of Googling. Those assistants answer using pages their crawlers are allowed to read. If GPTBot or ClaudeBot is blocked, your own site can’t inform the answer, and a competitor’s can.
Should I block AI crawlers or allow them?
It depends on the bot. Live-answer and AI-search bots (ChatGPT-User, OAI-SearchBot, Claude-User, PerplexityBot) should almost always be allowed. That’s how you get cited and recommended. Model-training crawlers (GPTBot, ClaudeBot, CCBot) are a separate choice: blocking them is a legitimate content/IP decision and does not remove you from live answers.
What is the “CDN blocking” check?
Cloudflare, Fastly and others ship a one-click “Block AI bots/scrapers” setting that refuses AI user-agents at the edge, silently overriding any “Allow” in your robots.txt. We probe your homepage with real bot user-agents and compare against a normal browser to flag when that’s happening.
Is this accurate / does it store my data?
We read your public robots.txt (the same file every crawler reads) and evaluate it with a pragmatic implementation of Google’s robots rules, then do three lightweight homepage requests. Nothing is stored. Treat it as a fast diagnostic, not legal advice.
Crawlable is step one. Being recommended is the goal.
Reputation Compass runs this automatically, tracking what ChatGPT and other AI engines actually say about you over time, alongside your reviews, Reddit, and social mentions, with human-reviewed responses.