tooluniverse-adverse-event-detection

Detect and analyze adverse drug event signals using FDA FAERS data, drug labels, disproportionality analysis (PRR, ROR, IC), and biomedical evidence. Generates quantitative safety signal scores (0-100) with evidence grading. Use for post-market surveillance, pharmacovigilance, drug safety assessment, adverse event investigation, and regulatory decision support.

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 content is highly actionable with concrete executable tool calls and a clear phased workflow, but it is over-long and monolithic, duplicating tool documentation inline rather than offloading it to reference files, and lacks explicit feedback-loop checkpoints for its batch operations.

Suggestions

Move the 'Tool Parameter Reference', 'Fallback Chains', and per-phase 'Output for Report' templates into separate reference files (e.g. TOOL_REFERENCE.md, REPORT_TEMPLATE.md) and link to them one level deep, reducing SKILL.md to an overview plus the phased workflow.

Add explicit validate->fix->retry checkpoints between phases (e.g., confirm FAERS returned data and apply the fallback chain before proceeding to disproportionality, and re-run any event whose response shape deviates from the documented structure).

Trim the fabricated sample numbers inside the output-for-report markdown blocks or mark them clearly as illustrative, and de-duplicate tools that already appear in the phased code examples to cut length without losing actionability.

Dimension	Reasoning	Score
Conciseness	The body is reference-grade and mostly non-redundant conceptually, but at ~1100 lines it duplicates tool coverage (full Tool Parameter Reference re-lists tools already shown per-phase) and carries large illustrative output-for-report templates with fabricated sample counts that could be trimmed, fitting the score-2 'mostly efficient but could be tightened' anchor rather than the lean score-3 anchor.	2 / 3
Actionability	It provides fully executable tu.tools.* calls with exact parameters and real response JSON shapes, a complete parameter reference table, fallback chains, and concrete error-handling code (the NOT_FOUND check), matching the score-3 'fully executable, copy-paste ready' anchor.	3 / 3
Workflow Clarity	The 10-phase pipeline with an ASCII diagram, numbered subsections, a per-phase completeness checklist, and NOT_FOUND error handling is clearly sequenced, but batch disproportionality operations lack explicit validate->fix->retry feedback loops and inter-phase verify-checkpoints, which the rubric caps at 2 for batch/destructive workflows.	2 / 3
Progressive Disclosure	No bundle files exist and the skill is a single monolithic ~1100-line SKILL.md holding all phases, the full tool reference, fallback chains, and edge cases inline; it is well-sectioned (avoiding a score-1 wall of text) but content that should be split into separate reference files is kept inline, matching the score-2 anchor.	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.

The description is specific, trigger-rich, complete, and clearly differentiated, with third-person voice and an explicit 'Use for' clause covering both capability and triggering context.

Dimension	Reasoning	Score
Specificity	The description enumerates concrete actions ('Detect and analyze adverse drug event signals', 'Generates quantitative safety signal scores (0-100) with evidence grading') alongside specific methods (PRR, ROR, IC) and data sources (FAERS, drug labels), matching the score-3 anchor of multiple specific concrete actions.	3 / 3
Completeness	It explicitly answers both 'what' (detect/analyze signals, generate 0-100 scores with evidence grading) and 'when' via an explicit 'Use for post-market surveillance, pharmacovigilance, ...' trigger clause, matching the score-3 anchor that requires both with explicit triggers.	3 / 3
Trigger Term Quality	It surfaces natural user-facing terms ('post-market surveillance', 'pharmacovigilance', 'drug safety assessment', 'adverse event investigation', 'regulatory decision support', 'FAERS') that a user would actually say, giving good coverage of common variations per the score-3 anchor.	3 / 3
Distinctiveness Conflict Risk	The niche is sharply defined by FAERS disproportionality analysis (PRR/ROR/IC) and a quantitative Safety Signal Score, and the body explicitly differentiates it from the sibling pharmacovigilance skill, making it unlikely to trigger for the wrong skill.	3 / 3
	Total	12 / 12 Passed

Validation

87%

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 — 14 / 16 Passed

Validation for skill structure

Criteria	Description	Result
skill_md_line_count	SKILL.md is long (1116 lines); consider splitting into references/ and linking	Warning
relative_links	Relative link issues: 1 missing	Warning

	Total	14 / 16 Passed

Repository: wu-yc/LabClaw
Path: skills/pharma/tooluniverse-adverse-event-detection/SKILL.md
Commit: df37802

Reviewed: 2 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.