ANSWERS

What is ClaudeBot?

ClaudeBot is Anthropic's web crawler that collects public web content which may be used to train future Claude models. It is separate from Claude-SearchBot, which handles search indexing, and Claude-User, which fetches pages when a person asks Claude about them. ye5's GEO Web audit checks whether your site's pages are accessible and readable to automated readers.

What it does with your content

ClaudeBot collects public content that may be used to train future Claude models. That collection is one behavior among three. Claude-SearchBot handles search indexing, and Claude-User handles fetching triggered by a person's question mid-conversation. Training, search indexing, and user-triggered fetching are different purposes, and access choices for one do not settle the others. Treat them as separate decisions when you set crawler rules for GEO.

What you can verify in your own logs

—Verify the request identity behind the claimed user agent, because a user-agent string can be copied.
—Record the paths requested, so you know which parts of the site automated readers reach.
—Record the response status codes returned, so you can see what was served and what was refused.
—Read requests as requests: they do not prove training, indexing, or citation.

Logs show requests and nothing more. A clean sitemap supports discovery of your URLs (see: How does my sitemap affect AI discovery?). What Claude says about your brand is a separate measurement from what your server recorded.

Should you allow it?

Allowing training collection is a content-policy choice, independent of search access, and it is not required for visibility (see: Should I block AI crawlers from my site?). Blocking does not erase existing datasets, and it does not prevent citations drawn from other sources. Knowing the crawler exists is not knowing what it can read. ye5's GEO Web audit checks your site's access and readability, and reports pattern-level fixes that scale.

FAQ

How do I see ClaudeBot in my logs?

Filter your server or CDN logs for the ClaudeBot user agent, or use a host that labels verified bots.

Does ClaudeBot execute JavaScript?

Capabilities vary by system, so do not assume every bot skips JavaScript. Keep essential facts in the initial HTML, which is the robust compatibility practice.