Synthvora Growth
AI Assistants

Perplexity Prompts

AI answer engine that searches the web and cites sources in its replies.

10 ready-to-use prompts

Researched for Perplexity real features. Replace the [BRACKETED] parts with your own details, copy, and paste.

1Cited Research Briefing
Research [TOPIC] and write a briefing for [AUDIENCE]. Requirements: every factual claim must carry an inline citation to a real source, prioritize primary sources and reporting from the last [TIME_WINDOW], and flag any claim where sources disagree. Structure: executive summary (5 bullets), background, current state with data points, competing viewpoints with named sources, and what is still unknown. End with [NUMBER] suggested follow-up questions I can ask to go deeper. Do not pad with generic filler; if evidence is thin, say so plainly.
Tip: Run this in Pro Search so Perplexity reasons over more sources before answering.
2Deep Research Report
Produce a long-form research report on [TOPIC] suitable for [USE_CASE, e.g. a board memo, a graduate seminar]. Use Deep Research mode and cover: historical background, current landscape with 2025-2026 data, key players and their positions, quantitative evidence in tables where useful, major debates with cited arguments on each side, and a forward-looking assessment for [TIME_HORIZON]. Every section needs citations. Target [WORD_COUNT] words. Conclude with an annotated list of the [NUMBER] most authoritative sources and one line on why each matters.
Tip: Deep Research takes several minutes; use it for reports, not quick questions.
3Claim Fact-Check
Fact-check this claim: '[CLAIM]'. Investigate using primary sources, official data, and reputable reporting. Deliver a verdict first: True, Mostly True, Misleading, Mostly False, or False, with a one-paragraph justification. Then show your work: a timeline of the relevant facts, quotes or data points from at least [NUMBER] independent sources with citations, and the strongest counter-argument to your verdict stated fairly. If the claim is too vague to verify, rewrite it into [NUMBER] testable versions first and check each one.
Tip: Paste viral claims here before resharing; ask for the verdict up front.
4Product Comparison with Sources
Compare [PRODUCT_A] vs [PRODUCT_B] (and [PRODUCT_C] if relevant) for [USE_CASE] on a [BUDGET] budget. Build a comparison table covering: price, key specs, strengths, weaknesses, and who each is best for. Every spec and price must be cited to a current source (manufacturer page, reputable review, or 2026 retailer listing). Add a 'what reviewers consistently say' section based on cited reviews, and finish with a clear recommendation for [USER_PROFILE] plus one runner-up. Flag any spec you could not verify instead of guessing.
Tip: Ask Perplexity to recheck prices if the answer is more than a few days old.
5Space-Based Project Research
I am organizing my research on [PROJECT] in a Perplexity Space. First, propose a Space structure: [NUMBER] named threads or file groups covering [SUBTOPICS], and what to upload into each. Then, for the first thread on [SUBTOPIC_1], write the opening research brief: objectives, key questions, and the sources I should prioritize. Keep everything organized so future threads build on earlier findings instead of repeating them. End with a checklist of what I should upload (files, links, notes) to make the Space genuinely useful over [TIME_HORIZON].
Tip: Create the Space first, upload your files, then run follow-ups inside it.
6Uploaded File Analysis
I have uploaded [FILE_DESCRIPTION, e.g. a 30-page industry report, lecture slides]. Analyze it and produce: (1) a 10-bullet summary of the essential points, (2) every statistic or data point with its page or section reference, (3) the document's main argument and its 3 strongest supporting pieces of evidence, (4) gaps, biases, or missing perspectives you notice, (5) [NUMBER] follow-up questions I can ask to extend this research beyond the document using web sources. Distinguish clearly between what the document says and what you add from outside knowledge, with citations for the latter.
Tip: Upload PDFs, slides, or images; Perplexity reads them and cites web sources for additions.
7Model Picker Showdown
I want to use Perplexity's model picker to answer [QUESTION] from multiple angles. First, tell me which [NUMBER] models I should select for this question (e.g. a reasoning model for analysis, a fast model for breadth) and what each one's answer will likely emphasize. Then give me the exact follow-up prompts to run under each model so the comparison is fair. Finally, provide a rubric with [NUMBER] criteria for judging which answer is best for [MY_GOAL], so I can pick a winner on substance instead of eyeballing it.
Tip: Pro subscribers can switch models mid-thread; compare before committing to one.
8Topic Learning Path
Teach me [TOPIC] from zero to working knowledge for [MY_GOAL]. Design a learning path of [NUMBER] stages, each with: the core concepts, one cited explanation I should read, and a self-test question. After presenting the path, teach me stage 1 now in plain language with a concrete example from [MY_FIELD]. End with [NUMBER] follow-up questions I can ask to continue to stage 2. Adjust depth to my level: [MY_LEVEL, e.g. beginner, practitioner]. Cite sources for every factual claim you make.
Tip: Use the related follow-up questions at the bottom to walk the path step by step.
9Literature-Style Evidence Review
Survey the available evidence on [QUESTION_OR_HYPOTHESIS] as if writing a mini literature review. Find and cite at least [NUMBER] credible sources (studies, official statistics, expert analyses). Organize by theme, note where sources agree and where they conflict, and rate the strength of evidence for each major claim as strong, moderate, or weak, with one line of reasoning per rating. Call out methodological limits. Conclude with what a reasonable person should believe given the current evidence and what would change that conclusion. No advocacy, just the evidence.
Tip: Ask for the evidence-strength labels explicitly; they keep answers honest.
10Event Timeline Reconstruction
Reconstruct a sourced timeline of [EVENT_OR_STORY] covering [DATE_RANGE]. For each key moment include: date, what happened in one sentence, and a citation to the source reporting it. Distinguish confirmed events from disputed ones, and note where early reporting was later corrected. Keep entries tight enough to scan in [NUMBER] minutes. End with [NUMBER] follow-up questions for angles the timeline does not cover (e.g. causes, key people, aftermath).
Tip: Timelines work best with a tight date range; widen it in follow-ups.
Back to all 214 platforms