tooluniverse-adverse-event-detection

Detect and analyze adverse drug event signals using FDA FAERS reports, drug labels, and disproportionality statistics (PRR, ROR, IC). Generates quantitative safety signal scores (0-100) with evidence grading. Use for post-market surveillance, pharmacovigilance, drug safety assessment, regulatory submissions, and detecting rare AE signals not visible in clinical trials.

Quality

70%

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

Fix and improve this skill with Tessl

tessl review fix ./plugin/skills/tooluniverse-adverse-event-detection/SKILL.md

Quality

Content

52%Weight 40%Scale 1-5

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

A well-structured, conceptually clear pipeline with strong sequencing and thresholds, but it is hampered by missing executable code and by referencing bundle files that do not exist, breaking progressive disclosure. Validation feedback loops are also more implicit than explicit for a multi-phase analytical workflow.

Suggestions

Add at least one concrete, executable Python snippet showing a disproportionality calculation or score computation, since the skill mandates running code rather than describing it.

Either provide the referenced PHASE_DETAILS.md, REPORT_TEMPLATE.md, TOOL_REFERENCE.md, and QUICK_START.md files or inline their essential content, so the references resolve to real bundle files.

Make inter-phase validation checkpoints explicit (e.g., 'only proceed to Phase 2 when Phase 1 returns >=3 reportable events') to add feedback loops for the batch analytical operations.

Dimension	Reasoning	Score
Conciseness	Mostly efficient with tight tool lists per phase and a clear workflow diagram, but several sections explain domain concepts Claude likely already knows (Naranjo algorithm steps, biological plausibility reasoning) that could be tightened or moved to references.	3 / 5
Actionability	Provides concrete tool names, signal thresholds (PRR>=2.0, N>=3, lower CI>1), and scoring component weights, but offers no executable code examples despite the 'COMPUTE, DON'T DESCRIBE' directive and defers most execution detail to PHASE_DETAILS.md.	3 / 5
Workflow Clarity	A well-sequenced 10-phase pipeline with a visual diagram, per-phase tool lists, and an explicit edge-case section; however, explicit validation checkpoints between phases are mostly implicit rather than enforced, and the report-first loop is stated as a principle rather than a checkpointed step.	4 / 5
Progressive Disclosure	The body references four companion files (PHASE_DETAILS.md, REPORT_TEMPLATE.md, TOOL_REFERENCE.md, QUICK_START.md) as 'in this directory', but none of these bundle files actually exist in the skill directory, leaving the references broken and the promised one-level-deep navigation unfulfilled.	2 / 5
	Total	12 / 20 Passed

Description

88%Weight 40%Scale 1-5

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, specific description that names concrete capabilities, data sources, and use cases with clear what/when coverage. Trigger terms are good and natural but could add a few synonyms; distinctiveness is solid though it does not explicitly distinguish from sibling pharmacovigilance skills in the description itself.

Suggestions

Add a brief clause distinguishing this skill from a general pharmacovigilance skill (the body contains this differentiation but the description frontmatter does not).

Include a few more natural synonyms such as 'side effects' or 'drug reactions' to broaden trigger coverage.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete actions (detect signals, disproportionality PRR/ROR/IC, quantitative Safety Signal Score 0-100, evidence grading) across named data sources (FAERS, FDA labels), giving comprehensive coverage.	5 / 5
Completeness	Explicitly states what the skill does (signal detection, scoring, evidence grading) and when to use it ('Use for post-market surveillance, pharmacovigilance, drug safety assessment, regulatory submissions'), answering both what and when concretely.	5 / 5
Trigger Term Quality	Includes natural user phrasings ('safety signals', 'adverse events', 'FAERS signals', 'post-market surveillance', 'pharmacovigilance') but lacks synonyms and the common drug-safety vernacular variations; missing a few natural terms.	4 / 5
Distinctiveness Conflict Risk	Occupies a clear, specialized niche (disproportionality-based signal detection with quantitative scoring) that is largely distinct; minor overlap risk with a general pharmacovigilance skill, which the description does not differentiate against.	4 / 5
	Total	18 / 20 Passed

Validation

100%

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 — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository: mims-harvard/ToolUniverse
Path: plugin/skills/tooluniverse-adverse-event-detection/SKILL.md
Commit: 7fe234c

Reviewed: 3 days 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.