Scientific Critical Thinking

This walks you through systematic evaluation of research methodology, experimental design, and statistical validity using frameworks like GRADE and Cochrane Risk of Bias. You get structured checklists for spotting biases (confirmation, selection, p-hacking), assessing confounders, and evaluating whether conclusions actually follow from the data. It's designed for reviewing papers, planning studies, or teaching critical analysis rather than writing formal peer reviews. The documentation includes detailed taxonomies of common biases and experimental design principles. Optionally integrates with scientific-schematics to generate diagrams like evidence assessment flowcharts, though that requires an OpenRouter API key and sends prompts to a third-party service.

Install to Claude Code

npx -y skills add k-dense-ai/scientific-agent-skills --skill scientific-critical-thinking --agent claude-code