Imbad0202/experiment-agent

This skill executes code experiments with real-time monitoring and manages human study workflows with ethics tracking. Its distinctive work begins afterward: it interprets statistical output, scans for 11 known fallacy patterns, and optionally re-runs the code to verify reproducibility. Designed for computational researchers, social scientists, statisticians, and methodologists who need not just a script runner but a systematic check that their results stand up to inspection. The planning mode uses Socratic dialogue to design experiments before execution, and all outputs use structured Markdown with Material Passports for pipeline integration.