tooluniverse-pharmacovigilance

Drug safety and adverse event analysis — FAERS spontaneous-report mining, FDA black-box warnings, signal detection (PRR, ROR, IC), risk factors by demographic/comorbidity, and label change tracking. Use for post-market safety surveillance, AE signal investigation, drug-AE association strength scoring, and pharmacovigilance reports.

Quality

82%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Low

Low-risk findings worth noting

Quality

Content

65%

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

The body is a richly actionable pharmacovigilance playbook with concrete tool usage and formulas, but it is verbose in places and leans on four bundle references that are not actually present, which weakens both workflow verification and progressive disclosure.

Suggestions

Create the referenced bundle files (SIGNAL_DETECTION.md, REPORT_TEMPLATES.md, CHECKLIST.md, TOOLS_REFERENCE.md) or remove the dead links so navigation works.

Trim explanations of well-known clinical concepts (Type A/B reactions, dechallenge/rechallenge, Naranjo algorithm) down to the pharmacovigilance-specific application, keeping only what Claude would not already know.

Add an explicit validation/retry loop for the report-first workflow — e.g., a concrete 'verify all phases pass CHECKLIST before finalizing the report' gate with error-recovery steps.

Dimension	Reasoning	Score
Conciseness	Mostly efficient and tool-specific, but spends tokens explaining concepts Claude already knows — on-target/off-target definitions, Type A/B reactions, dechallenge/rechallenge, the Naranjo algorithm, and what SJS/TEN/DRESS are — which could be tightened.	2 / 3
Actionability	Highly executable: exact tool calls with correct parameter names, a WRONG/CORRECT parameter table, the PRR formula, a signal-score formula, MedDRA British-spelling table, and an explicit instruction to run Python via Bash and report real results.	3 / 3
Workflow Clarity	Phases 0–7 are clearly sequenced with signal thresholds (PRR>2.0/>3.0) and a cross-check-against-mechanism step, but there are no explicit validate-then-proceed feedback loops, and the referenced CHECKLIST.md (the verification step) does not exist.	2 / 3
Progressive Disclosure	References are signaled one level deep (SIGNAL_DETECTION.md, REPORT_TEMPLATES.md, CHECKLIST.md, TOOLS_REFERENCE.md) but none of these files exist in the bundle, so navigation is broken; meanwhile the long inline Reasoning Strategies and Tool Parameter Reference sections could themselves be split out.	2 / 3
	Total	9 / 12 Passed

Description

100%

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

A strong, well-scoped description that pairs a concrete capability list with an explicit 'Use for' trigger clause in third person. It clearly distinguishes the skill's pharmacovigilance niche with minimal fluff.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete actions — "FAERS spontaneous-report mining", "FDA black-box warnings", "signal detection (PRR, ROR, IC)", "risk factors by demographic/comorbidity", "label change tracking" — rather than vague language.	3 / 3
Completeness	Explicitly answers both what (drug safety / adverse-event analysis with named methods) and when ("Use for post-market safety surveillance, AE signal investigation...") with an explicit trigger clause.	3 / 3
Trigger Term Quality	The "Use for" clause surfaces natural domain phrasings users would say — "post-market safety surveillance", "AE signal investigation", "drug-AE association strength scoring", "pharmacovigilance reports" — giving good coverage of likely trigger terms.	3 / 3
Distinctiveness Conflict Risk	The pharmacovigilance / post-market drug-safety niche is narrow and well-scoped, with triggers unlikely to fire for unrelated skills.	3 / 3
	Total	12 / 12 Passed

Validation

93%

Warnings & errors only

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 15 / 16 Passed

Validation for skill structure

Criteria	Description	Result
relative_links	Relative link issues: 4 missing	Warning

	Total	15 / 16 Passed

Repository: mims-harvard/ToolUniverse
Path: plugin/skills/tooluniverse-pharmacovigilance/SKILL.md
Commit: 089eb8e

Reviewed: about 9 hours ago

Table of Contents

Discovery Implementation Validation

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.