Claude-User vs PerplexityBot

Claude-User and PerplexityBot compared on 10 documented fields, from 19 claims, last verified 2026-09-13.

Field	Claude-User	PerplexityBot	Comparison
cites_sources	not_documented	yes	different
honours_crawl_delay	yes	not_documented	different
publishes_ip_list	yes	yes	same
purpose_documented	user_fetch	search	different
reads_llms_txt	not_documented	not_documented	same
respects_robots_txt	yes	yes	same
separate_search_and_training_tokens	not_documented	not_documented	same
user_agent_string_full	not_documented	Mozilla/5.0 AppleWebKit/537.36 (KHTML, like Gecko; compatible; PerplexityBot/1.0; +https://perplexity.ai/perplexitybot)	different
user_fetch_exception_documented	no	not documented	different
verifiable_by_rdns	not_documented	not_documented	same

Differences

Field	Claude-User statement	PerplexityBot statement
cites_sources	Anthropic does not publish whether Claude responses cite pages fetched by Claude-User.	Perplexity links to the websites that PerplexityBot surfaces in Perplexity search results.
honours_crawl_delay	Claude-User respects the Crawl-delay directive in robots.txt when fetching pages.	Perplexity does not publish whether PerplexityBot honours the robots.txt Crawl-delay directive.
purpose_documented	Claude-User accesses websites in response to questions that individual users ask Claude.	PerplexityBot surfaces and links websites in Perplexity search results.
user_agent_string_full	Anthropic does not publish a full user agent string for Claude-User.	PerplexityBot sends the user agent string Mozilla/5.0 AppleWebKit/537.36 (KHTML, like Gecko; compatible; PerplexityBot/1.0; +https://perplexity.ai/perplexitybot).
user_fetch_exception_documented	Anthropic applies the same robots.txt statement to Claude-User as to ClaudeBot and Claude-SearchBot and documents no exception for user-initiated fetches.	not documented