Engine guide: Claude

How to get cited by Claude

Claude may look at a different index from Google or Bing based assistants. A site well covered by those can be thin where Claude looks.

The short answer

Anthropic documents three agents: ClaudeBot for training, Claude-SearchBot to index content for search and Claude-User for fetches a person asks for. All honour robots.txt. Third-party testing found Claude's web search results matched Brave Search, so check your pages there too.

What ranked on 26 September 2026: Anthropic's crawler help page and trade coverage of the three-bot split rank here. The Brave link rests on a subprocessor list and testing, not an announcement.

Last reviewed 19 October 2026 · 3 sources

What is specific to Claude

Two things set it apart: the bot split and the likely index behind its search.

Blocking ClaudeBot does not stop Claude-SearchBot or Claude-User
Blocking Claude-SearchBot or Claude-User may reduce visibility
Brave appeared on Anthropic's subprocessor list on 19 March 2025
In one test Claude's ten cited results matched Brave's for the same query
Claude cited in 55% of responses, with about 13 sources when it did

Claim and reality

Things that sound right and are not.

Flip a card for what is actually true.

Claim
A blanket AI-bot block is the safe default.
Claim
If Google ranks us, Claude will find us.
Claim
Claude will cite our blog for everything.

Where to start

In order. Tick as you go, or copy the list.

0 of 5 done

Audit robots.txt and CDN rules for ClaudeBot, Claude-SearchBot and Claude-User. Set each on purpose.
Search your top ten buyer questions on search.brave.com. Note whether your pages appear.
If key pages are missing, link them from pages Brave does index.
Put supporting facts in short, attributable sentences.
Test the same questions in Claude with web search on. Record which URLs are linked.

Questions

Asked most.

Does blocking ClaudeBot affect citations?

Anthropic describes the agents as separate. Blocking ClaudeBot alone covers training, not search or user fetches.

Does Claude use Google?

I found no statement saying so. Testing and Anthropic's subprocessor list point to Brave.

Can I submit my site to Brave?

Inclusion depends on Brave's own crawl. Solid internal and external links are the dependable route.

Evidence, labelled

What is confirmed, what is reported, what is a hunch.

Each claim is labelled and sourced. Google rows speak only for Google.

Last reviewed 19 October 2026 · 3 sources

Confirmed: Stated in the platform's own documentation or measured in published data.
Reported: Third-party study or vendor data. Method and conflicts noted.

Anthropic runs three separate agents.
Confirmed
ClaudeBot (training), Claude-User (user fetch), Claude-SearchBot (search index). Updated 7 April 2026.
Anthropic, crawlers (opens in a new tab)
Claude's web search uses Brave.
Reported
Subprocessor list entry dated 19 March 2025 and a test where cited results matched Brave's exactly. Anthropic did not announce it there.
Willison, Claude and Brave (opens in a new tab)
Claude cites often but selectively.
Reported
55% of responses carried citations, with about 13 sources each. A May 2026 vendor study of 25M+ links.
Muck Rack, May 2026 (opens in a new tab)

Sources, fetched 19 October 2026

01Does Anthropic crawl data from the web, and how can site owners block the crawler? (opens in a new tab)Anthropic Help Center · Updated 7 Apr 2026
02Anthropic appears to be using Brave to power web searches for Claude (opens in a new tab)Simon Willison's Weblog · 21 Mar 2025
03Earned media still drives 84% of AI citations (What Is AI Reading?, May 2026) (opens in a new tab)Muck Rack · Updated 7 May 2026

Go deeper

Want this applied to your own site?

Send me your site and what you have tried. I will say which step comes first for you and which you can skip.

Send me your siteWritten reply within 24 hours.