Tooluniverse Adverse Event Detection

This is a serious pharmacovigilance tool that runs FDA FAERS disproportionality analysis to detect adverse drug event signals. It calculates PRR, ROR, and information component statistics with confidence intervals, then produces a quantitative safety signal score from 0 to 100. The workflow pulls FAERS reports, mines FDA labels for boxed warnings, cross-references OpenTargets and DrugBank for mechanism-based context, and applies the Naranjo causality algorithm. It's built for post-market surveillance and regulatory work where you need statistical rigor, not just anecdotal safety info. The skill emphasizes computation over description and uses multi-source triangulation to separate drug-specific signals from class-wide effects or confounders.

Install to Claude Code

npx -y skills add mims-harvard/tooluniverse --skill tooluniverse-adverse-event-detection --agent claude-code