Claude-User is Anthropic's user-action fetcher. It is not a bulk crawler and not a training crawler: it retrieves a page in real time when a Claude user's action needs live content from it. Its activity is driven by genuine user requests, not scheduled indexing.
Blocking Claude-User prevents these user-initiated fetches, so Claude can't pull your live page into a conversation even when a user explicitly asks about it.
How to check and control it
To check, look in your robots.txt for a rule targeting this user-agent. To allow or block it:
User-agent: Claude-User
Disallow: / # blocks it — remove or set "Allow: /" to permit
For an instant check across your whole site, use SEO AEO Specialist's free AI crawler checker, and see the llms.txt & AI crawler guide.
FAQ
Is Claude-User the same as ClaudeBot?
No. ClaudeBot crawls the web to train Claude models, while Claude-User fetches a single page in real time when a user's action requires it. Blocking Claude-User stops on-demand fetches; blocking ClaudeBot opts you out of model training. They are separate robots.txt rules.
Should I block Claude-User?
Usually not, if you want Claude to access your live pages when users ask about them. Because it acts on real user intent, blocking it mainly limits your reach. SEO AEO Specialist checks every Anthropic user-agent so you can decide deliberately rather than by accident.
See where you stand
SEO AEO Specialist runs a free AI-visibility audit and hands you the exact fixes. One-off report €9, unlimited €19/mo.
Run your free audit →