--- language: en title: "PerplexityBot" description: "PerplexityBot is Perplexity’s search crawler. Learn its purpose, how to check permissions, and why access does not prove citation." canonical: "https://tropk.ai/en/resources/glossary/perplexitybot" last_modified: "2026-10-12T16:11:26.708Z" --- # PerplexityBot Reviewed on 2026-10-06. ## What it means PerplexityBot is [Perplexity](/en/resources/glossary/perplexity)’s [crawler](/en/resources/glossary/crawler), designed to find and present websites, with links, in the service’s search results. The company states that it does not collect content to train foundational AI models. The [User-Agent header](/en/resources/glossary/user-agent) declares the agent’s name but does not authenticate its origin. The operator publishes its own IPs in a list separate from that of [Perplexity-User](/en/resources/glossary/perplexity-user). ## Why it matters for marketing A content team may want to make pages easier for Perplexity to discover without interpreting every bot visit as data collection for training. This purpose helps define permissions in [robots.txt](/en/resources/glossary/robots-txt) and the site’s security rules. It also prevents crawl logs from being presented to clients as evidence of citations or new visitors. ## Example in practice Consider an agency that publishes comparisons of construction materials. Its robots.txt allows PerplexityBot, but the firewall report shows requests from this agent receiving HTTP 403. After checking the origin against the official list, the team decides to review the rule blocking these visits. The observation reveals an access issue, not a proven loss of citations. ## How to apply or verify Check PerplexityBot’s permissions in robots.txt and how it is handled by the firewall. The documentation recommends combining the User-Agent with the official IPs in access rules, using the operator’s current list. Record the URL and status before and after the change; a simulated request tests the rule but does not prove identity. ## Important distinctions PerplexityBot and Perplexity-User serve different mechanisms: search [crawling](/en/resources/glossary/crawling) and [retrieval](/en/resources/glossary/information-retrieval) at people’s request. The note about ignoring robots.txt is documented for Perplexity-User and should not be applied to the search bot. Allowing PerplexityBot facilitates the access recommended by the provider, but does not guarantee a page’s inclusion, ranking, [citation](/en/resources/glossary/ai-citation), or traffic. ## Sources and references - [Perplexity Crawlers]() — Perplexity ## Continue a leitura - [Perplexity-User]() - [User agent]() - [robots.txt]() ## About the author [Vinicius Pacheco]() Responsible for Tropk’s technical content, focusing on crawling, indexing, structured data, and agent access to content. --- [All pages in Markdown]()