What problem does it solve? Researchers, reviewers, and analysts often need to judge whether a scientific claim is actually supported by its evidence. This Skill provides a systematic framework for critiquing methodology, detecting biases, evaluating statistics, and grading evidence quality instead of relying on intuition. ## Core Features & Use Cases - Methodology and Design Critique: Assess internal, external, construct, and statistical conclusion validity, including randomization, blinding, controls, and confounding. - Bias and Fallacy Detection: Identify cognitive, selection, measurement, and analysis biases (p-hacking, HARKing, publication bias) plus 40 named logical fallacies with detection strategies. - Evidence Grading: Apply the GRADE system, Cochrane Risk of Bias, ROBINS-I, and the traditional evidence hierarchy to rate confidence in findings. - Use Case: When reviewing a preprint claiming a supplement improves memory, use this Skill to check the study's power, control group adequacy, selective reporting, and whether causal language is justified by the correlational design. ## Quick Start Ask the AI to critically evaluate the methodology and evidence quality of an attached research paper using the scientific-critical-thinking skill.
