Consensus Prompts

Ask Consensus the yes/no question: '[RESEARCH_QUESTION]' and read the Consensus Meter first to see how the 200M+ paper corpus splits between Yes, No, and Possibly. Then open Study Snapshots for the [NUMBER] most-cited papers behind the verdict and compare their sample sizes, study types, and publication dates. Summarize the balance of evidence in about 150 words, note whether RCTs or observational studies dominate each side, and name the single strongest paper supporting each answer.

Use Consensus Deep Search on the topic '[LITERATURE_TOPIC]' and let it synthesize 50 to 200 relevant papers into a draft literature review. From the synthesis, extract the three most agreed-upon findings, the two biggest open disagreements, and the [NUMBER] most important citations for a [ACADEMIC_LEVEL] audience. Then filter by study type to check whether high-quality journals change any conclusion, and rewrite the review's introduction paragraph with proper scientific caution.

I need an evidence-weighted brief on '[DECISION_TOPIC]' for [AUDIENCE]. Use Consensus to find peer-reviewed research on each option I am considering, apply the Consensus Meter to judge agreement across studies, and weight findings by sample size and study quality. Present a one-page brief: what the evidence supports, where studies conflict, the effect sizes that matter for [CONTEXT], and three open questions that weaker studies have not answered yet.

Fact-check the viral claim '[VIRAL_CLAIM]' using Consensus. Search the claim in quotes, check the Consensus Meter split across the paper corpus, and turn on Medical Mode if it is a health claim. Open Study Snapshots for the papers the Meter leans on, verify they actually studied this claim and not a related one, and write a myth-bust summary for [AUDIENCE]: verdict, what the studies really measured, and the nuance the viral version leaves out.

Compare how peer-reviewed studies disagree on '[DEBATED_TOPIC]' using Consensus. Find [NUMBER] studies that reached opposing conclusions, then for each one record from its Study Snapshot: methodology, sample size, population, journal, and year. Write a comparison table in prose explaining why the disagreement exists: different methods, different populations, or different time periods. Conclude with which side has the stronger methodology for [USE_CASE] and why.

Use Consensus to investigate '[RESEARCH_NICHE]' purely through methodology. Filter results by study type to separate RCTs, cohort studies, and meta-analyses, then read the Study Snapshots of the [NUMBER] highest-quality papers to note their designs, sample sizes, and limitations. Summarize which methods dominate this niche, what their known blind spots are, and which study design a researcher in [FIELD] should trust most for this question.

Pick the most influential paper Consensus found on '[PAPER_TOPIC]' and open it in Ask Paper chat. Ask it to explain the paper's core hypothesis, walk through the methodology step by step, report the exact sample size and key statistics, list the limitations section's main points, and explain how the authors interpreted their own results. Then have it compare those claims to the broader Consensus Meter verdict so I can see if this paper is typical or an outlier for [FIELD].

Launch the Consensus Research Agent on '[EMERGING_TOPIC]', letting it chain search, citation crawling, and study comparison automatically. Instruct it to: map the key papers and how they cite each other, identify the Graph (beta) citation clusters, summarize where agreement is forming, and flag the newest [NUMBER] papers that shift the conversation. Compile the output into a research map with three clusters, the central paper in each, and one unanswered question per cluster for [AUDIENCE].

In Medical Mode, evaluate the claim that '[TREATMENT_CLAIM]' for [CONDITION]. Check the Consensus Meter across clinical literature, filter for RCTs and meta-analyses only, and read Study Snapshots for the [NUMBER] largest trials to extract dosages, durations, side effects, and effect sizes. Write a plain-language summary for [AUDIENCE]: what the evidence supports, what it does not, the risks reported, and what to discuss with a clinician.

Run an evidence scan on '[SCANNED_TOPIC]' with strict filters: top-quartile journals only, [MIN_SAMPLE_SIZE] minimum sample size, and papers from [START_YEAR] onward. Use the Consensus Meter on the filtered set, then have the Copilot Summary synthesize the top papers. Compare this filtered verdict to the unfiltered one and report how much the conclusion changes, which weak studies the filters removed, and whether the remaining evidence is strong enough to act on for [PURPOSE].