k
kritak
back to agents
?ResearchmodeFreemium

Perplexity

by Perplexity AI

A search-first research agent that reads, synthesizes, and cites every claim — replacing "Google then read 8 tabs" with one structured answer.

Notable for
Made citations a first-class feature instead of a footnote, raising the bar across the whole research-agent category.

$ cat curator-note.md

Perplexity's strength is the discipline of citation-first answers. Every claim gets a numbered source you can click and verify, which sounds modest until you've spent a few weeks using it and notice you no longer feel the urge to double-check obvious facts. Pro Search and Deep Research go further — they plan the investigation, run multiple queries in parallel, and produce a structured writeup with sections and inline references. For "I need to understand a topic by tonight" tasks, it's hard to beat.

Where it falls short is depth. Perplexity is excellent at synthesizing the consensus view of a topic from public web sources, but it struggles with anything that requires real expertise — niche academic disputes, contested industry topics, or questions where the right answer isn't on the first page of Google. It also has a tendency to confidently cite low-quality sources alongside good ones; the citation count goes up, the actual rigor stays mixed. Treat it as a strong starting point, not a final answer.

Use Perplexity when you need to get up to speed on an unfamiliar topic quickly, prep for a meeting, or write something that needs to be plausibly sourced rather than originally researched. For deeper work — academic literature reviews, due diligence on a specific company, anything where the answer matters — pair it with traditional research and human judgment. As a "first 80 percent" tool it's exceptional. As a final source it's not.