Critically analyze [TEXT OR CLAIM] for [PURPOSE AND AUDIENCE]. Use this context about the source, setting, and stakes: [SOURCE AND CONTEXT]. Evaluate it against [EVALUATION CRITERIA], using only the text and evidence I provide: [AVAILABLE EVIDENCE OR NOTES]. My required scope and length are [SCOPE AND WORD COUNT]. Start by state the central claim in one neutral sentence and identify the argument's key subclaims. Then produce a structured analysis with these sections: 1. Thesis and purpose: what the author is trying to establish, for whom, and through what framing. 2. Reasoning map: the major claims, supporting reasons, evidence offered, and any assumptions needed for the conclusion to follow. 3. Evidence assessment: relevance, sufficiency, source quality, date, sample or method limits, and the difference between correlation, causation, anecdote, and representative evidence where applicable. 4. Critical evaluation: strongest aspects, specific weaknesses, omissions, alternative explanations, counterarguments, and potential bias or incentive. Discuss only issues supported by the material; do not call something biased merely because you disagree with it. 5. Conclusion: a calibrated judgment that states what the material supports, what remains uncertain, and what additional evidence would change the assessment. Quote sparingly and cite passage locations, page numbers, or note labels when available. Do not invent sources, facts, author intentions, or citations. Distinguish clearly among what the source says, what follows logically, and your evaluation. Before answering, check that the conclusion does not overreach the evidence and that every criticism names the exact claim, evidence gap, or reasoning step at issue. Ask up to 3 clarifying questions only if a required input is missing.
Fill in
| Placeholder | What to enter | Example |
|---|---|---|
| [TEXT OR CLAIM] | Paste the text, excerpt, argument, or specific claim you want analyzed. | “A four-day workweek will increase productivity in every knowledge-work organization because employees will be more rested.” |
| [PURPOSE AND AUDIENCE] | State why you need the analysis and who will read it. | A 900-word analysis for an MBA course discussion on evaluating management claims. |
| [SOURCE AND CONTEXT] | Describe the author, publication, date, setting, and any relevant background you know. | The claim appears in a 2025 opinion column by a workplace consultant; the column cites two company case studies but gives no methods. |
| [EVALUATION CRITERIA] | List the standards that matter, such as methodological rigor, policy feasibility, ethical implications, or rhetorical effectiveness. | Logical validity, quality of evidence, generalizability, causal inference, implementation conditions, and counterarguments. |
| [AVAILABLE EVIDENCE OR NOTES] | Provide any sources, data, annotations, or notes that may be used in the evaluation. | My notes summarize two pilots: one 32-person marketing agency and one 280-person software company; neither used a control group. |
| [SCOPE AND WORD COUNT] | Specify the focus, required sections, and approximate target length. | Focus on the universal claim and causal reasoning; 850–950 words with section labels. |
How to use
- Paste the complete text or a clearly bounded excerpt, then add source context and the standards by which it should be judged.
- Include source notes and citations you already have; do not ask the model to fill missing research with plausible references.
- Check that each major criticism points to a passage, inference, method limitation, or omitted comparison rather than a preference.
- Follow up with: “Now turn the reasoning map into a two-column table of claims, evidence, assumptions, and the question each assumption raises.”
Variations
Academic article critique
Use this to assess a research paper's method and conclusions.
Critique this research article: [ARTICLE OR EXCERPT]. Assess [RESEARCH QUESTION], study design, sample, measures, analysis, limitations, and whether the conclusions match the results. My course or field standards are [STANDARDS], and relevant context is [CONTEXT]. Write [WORD COUNT] words with a method summary, strengths, limitations, validity threats, generalizability, and a calibrated conclusion. Cite page or section locations from the material. Do not invent studies or statistical results. Flag information needed for a full assessment and ask up to 3 questions only if required.
Policy proposal analysis
Use this to test a proposed policy before recommending it.
Analyze this policy proposal: [PROPOSAL]. The jurisdiction, affected groups, and implementation context are [CONTEXT]. Evaluate it for [CRITERIA, SUCH AS FEASIBILITY, COST, EQUITY, LEGAL AUTHORITY, AND ENFORCEMENT]. Produce a decision memo with the proposal's theory of change, assumptions, stakeholder effects, implementation dependencies, risks, alternatives, and questions requiring evidence or legal review. Separate documented facts from forecasts. Do not estimate costs or legal outcomes without supplied sources. Ask up to 3 questions only if essential information is missing.
Argument revision plan
Use this after drafting an argument and before rewriting it.
Diagnose this draft argument: [DRAFT]. Its intended conclusion and audience are [GOAL AND AUDIENCE], and the evidence I am allowed to use is [EVIDENCE]. Return an argument map, the three most consequential reasoning gaps, unsupported or overstated sentences, missing counterarguments, and a prioritized revision plan. Then rewrite only the thesis and outline, not the full draft. Preserve claims that the evidence can support and mark claims that require a source. Check that suggested changes do not introduce new facts. Ask up to 3 questions only if needed.
Tips
- Analyze the bridge between evidence and conclusion, not only the quality of individual sources; many weak arguments fail at an unstated assumption.
- A limitation is meaningful only when you explain what it could change: confidence, generalizability, causality, magnitude, or applicability.
- Treat an author's incentive or framing as one piece of context, not proof that the claim is false.
- Use exact citations or passage labels while drafting; this prevents a critique from becoming a list of unsupported impressions.
FAQ
Is critical analysis the same as summarizing?
No. A summary reports what a source says; critical analysis assesses how well its claims are supported and what limits apply.
Can AI find flaws in an argument reliably?
It can surface assumptions and gaps, but it may miss domain context or misread evidence. Verify its observations against the original material and primary sources.
Should a critical analysis always be negative?
No. A sound analysis identifies the strongest evidence and reasoning as well as limitations, then reaches a proportionate conclusion.