Perplexity · Answers
Perplexity-User
Perplexity's public argument is that this is not a crawler at all: "User-driven agents only act when users make specific requests, and they only fetch the content needed to fulfill those requests. This is the fundamental difference between a user agent and a bot." Mechanically that holds: the docs say it "supports user actions within Perplexity," visiting a page "when users ask Perplexity a question" in order to "include a link to the page in its response," and it feeds neither index nor model: "It is not used for web crawling or to collect content for training AI foundation models." Verification tops out at an IP match against a four-prefix file dated October 2025. Four declared prefixes is a narrow fleet for a live consumer surface, so strict range checking here is high-precision and low-recall by construction.
- Operated by
- Perplexity
- Purpose
- Answers
- robots.txt token
- Perplexity-User
- Verification
- IP verified
Fetches a page in real time to answer a question someone just asked.
How to verify Perplexity-User
Perplexity publishes the IP ranges Perplexity-User crawls from, and we fetch that list on a schedule. A request claiming to be Perplexity-User can therefore be checked against 4 published ranges: one that does not match is not this crawler.
Published ranges last confirmed by us on .
User agent
Perplexity publishes this user agent for Perplexity-User. Match on the Perplexity-User product token rather than the whole string: vendors revise the surrounding version and URL fragments without notice.
Mozilla/5.0 AppleWebKit/537.36 (KHTML, like Gecko; compatible; Perplexity-User/1.0; +https://perplexity.ai/perplexity-user)robots.txt for Perplexity-User
Perplexity's documentation is inconsistent on this point, and the inconsistency sits inside a single table cell. One sentence reads: "Since a user requested the fetch, this fetcher generally ignores robots.txt rules." The other reads: "`Perplexity-User` controls which sites these user requests can be made to." Those cannot both hold (a token cannot be the control surface and be outside its own control), and Perplexity does not reconcile them. Read the literal one and note that "generally" is load-bearing: no criteria are published for when the directive is honoured. Do not treat a `Disallow` as a reliable stop for this token; the only mechanism supported by Perplexity's own documentation is a network or WAF block.
Block
User-agent: Perplexity-User
Disallow: /Allow
User-agent: Perplexity-User
Allow: /robots.txt is a request, not an enforcement mechanism. It is honoured by convention, and a crawler that ignores it is stopped at your edge, not in a text file.
What blocking Perplexity-User costs you
Start from what a robots directive cannot buy you: on Perplexity's own account this fetcher generally ignores it, so the real decision is whether to deny at the network layer, and that is where the cost lands. Deny it and the live-fetch path is severed at the moment a human is asking Perplexity about your page: the fetch that the docs say produces "a link to the page in its response" simply fails, and the citation never appears. Nothing in your logs marks the loss: no error, no bounce, no referrer, just an answer written without you. Since these requests arrive one per question rather than on a sweep, their volume is closer to a demand signal for your content than to a crawl-budget line item.