Promptwatch Logo

Claude-User

Claude-User supports Claude AI users. When individuals ask questions to Claude, it may access websites using a Claude-User agent.
AnthropicClaude-User
AI AssistantUser Initiated

What is Claude-User?

Claude-User is the identity Anthropic uses when Claude retrieves a webpage at a user's direction. A request can follow a question or another task in Claude that requires current content from a site. It is different from Claude-SearchBot, which supports search indexing, and ClaudeBot, which collects material that could be used for model training.

Because the fetch is tied to a user request, Claude-User normally visits a specific URL needed for that interaction rather than sweeping a site to build a general index. The page may then inform Claude's answer to that user. Seeing this agent in a log is evidence of user-directed retrieval, not proof that Anthropic is crawling the domain broadly.

Allowing Claude-User can make content available when Claude responds to a user-directed web query. Blocking it can prevent retrieval at that moment and may reduce visibility in those responses. This control is separate from model training: Anthropic assigns potential training collection to ClaudeBot, so a Claude-User visit alone does not mean the page entered a training dataset.

Anthropic says its bots aim to crawl with minimal disruption, honor standard robots.txt directives, and respect Crawl-delay where appropriate. The records used for this page nevertheless classify Claude-User's robots support as partial, while Cloudflare's directory marks it as not following robots.txt. Site owners who need a hard block should verify the observed behavior and enforce access at the server or edge rather than rely on a directive alone.

The stable token is Claude-User. Cloudflare's bot directory records two browser-style user agents, differing only in the capitalization of the contact address: Mozilla/5.0 AppleWebKit/537.36 (KHTML, like Gecko; compatible; Claude-User/1.0; [email protected]) and Mozilla/5.0 AppleWebKit/537.36 (KHTML, like Gecko; compatible; Claude-User/1.0; [email protected]). This text is useful for log classification, but a user-agent header can be copied.

Anthropic also publishes source IP information for its crawlers, and Claude-User is marked as IP-verifiable. Check the source address as well as the header before treating a request as genuine. There is no Web Bot Auth signature directory listed for Claude-User.

Relevant for AI searchClaude

Is Claude-User relevant for AI search?

Yes. Claude-User feeds Claude, so the pages it can reach shape what those AI products say about you.

Claude-User crawls and indexes pages so the AI search or assistant behind it can retrieve them at answer time. A page it has never fetched cannot be quoted, summarized, or linked in that product's answers, so most sites keep it allowed to stay citable. Blocking it removes your pages from that AI surface and hands those citations to competitors.

Track Claude-User with Promptwatch

Promptwatch classifies Claude-User (Claude-User) in real time from your server and CDN logs. See exactly when it visits, which pages it requests, the status codes it gets, and how those crawls map to AI citations. Requests are validated against the operator's published IP ranges to filter out spoofed user agents.

How to handle Claude-User

Treat Claude-User as a user-directed retrieval agent. Leave it reachable if you want Claude to fetch your public pages in response to user queries. Restrict it if those requests are unwanted, but make that decision separately from your policy for ClaudeBot and model training.

Use this rule to state that Claude-User should not crawl the site:

User-agent: Claude-User
Disallow: /

Anthropic says this rule prevents its system from retrieving content for a user query. Because the available records disagree on full robots.txt compliance, check subsequent requests in your logs. If denial must be guaranteed, apply a server or edge rule and validate source IPs against Anthropic's published crawler information.

Examples

  • A documentation team sees Claude-User request a newly published API page after a developer asks Claude about that endpoint.
  • A publisher allows Claude-User but blocks ClaudeBot, permitting user-directed answers without granting the same preference for future training collection.
  • A security engineer checks the source address of a Claude-User request instead of trusting the copied user-agent text.

Frequently asked questions about Claude-User

Learn about AI visibility monitoring and how Promptwatch helps your brand succeed in AI search.

Claude-User is used when a person asks Claude a question or gives it a task that requires content from a website. It is a user-directed fetch rather than the identity Anthropic uses for broad training collection.

Related bots

Claude-SearchBot

Claude-SearchBot is Anthropic's crawler for improving search results shown to Claude users.

AI Assistant

ClaudeBot

ClaudeBot is Anthropic's crawler for gathering public web content that could contribute to the development of its generative AI models.

AI Crawler

Amazon Kendra

Amazon Kendra is a managed information retrieval and intelligent search service that uses natural language processing and advanced deep learning model.

AI Assistant

Amazon Q

Amazon Q Business is a generative artificial intelligence (generative AI)-powered assistant that you can tailor to your business needs.

AI Assistant

ChatGPT-Operator

Handles user-initiated requests from ChatGPT operator accessing external content; not used for automated crawling or AI training.

AI Assistant

Be the brand AI recommends

Monitor your brand's visibility across ChatGPT, Claude, Perplexity, and Gemini. Get actionable insights and create content that gets cited by AI search engines.

Promptwatch Dashboard