tooluniverse-clinical-data-integration

End-to-end drug safety review integrating FDA labels, FAERS adverse event reports, PRR/ROR disproportionality, pharmacogenomic biomarkers, clinical trial data, and published literature. Use for regulatory drug safety reviews, comprehensive pharmacovigilance reports, label-vs-real-world AE comparison, and clinical decision support for drug safety.

Quality

73%

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-clinical-data-integration/SKILL.md

Quality

Content

62%

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

A well-organized, clearly sequenced drug-safety workflow with concrete tool schemas and explicit checkpoints. It loses points for instructional padding, the absence of executable compute examples despite directing their use, and a monolithic structure with no reference splitting.

Suggestions

Add at least one runnable Python example for the 'COMPUTE, DON'T DESCRIBE' directive — e.g. a pandas/statsmodels snippet computing PRR/ROR with 95% CI from a FAERS count matrix.

Trim tutorial asides Claude already knows (missing-data interpretation, the bias-type enumeration) to recover token budget.

Move the per-phase tool reference (inputs/outputs) and the PRR/ROR/IC interpretation tables into a reference file referenced one level deep, keeping SKILL.md as a lean overview.

Dimension	Reasoning	Score
Conciseness	Mostly efficient and well-structured, but includes tutorial-style asides Claude likely already knows (e.g. 'Missing data is informative — a missing lab value may mean the test wasn't ordered', and the channeling/notoriety/protopathic bias enumeration) that could be trimmed.	2 / 3
Actionability	Tool names with exact input/output schemas and numeric signal thresholds are highly concrete, yet despite a 'COMPUTE, DON'T DESCRIBE' directive the body provides no executable Python code examples — guidance is actionable at the tool level but incomplete at the compute level.	2 / 3
Workflow Clarity	A clearly sequenced 0-6 phase pipeline with explicit checkpoints — signal criteria (PRR >= 2.0, lower CI > 1.0, N >= 3), a 5-step signal-credibility checklist, and evidence grading tiers — giving an unambiguous execution path.	3 / 3
Progressive Disclosure	Sections are well organized, but the skill is a monolithic single file with no bundle files (references/, scripts/, assets/ absent) and no one-level-deep references to split out the large tool reference and interpretation tables.	2 / 3
	Total	9 / 12 Passed

Description

85%

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 clearly conveys both capabilities and trigger contexts with good distinctiveness. The only weakness is trigger-term jargon that not every user would voice naturally.

Suggestions

Soften specialist jargon in the trigger clause — add plain-language phrasings like 'drug side effects', 'bad reactions to a medication', or 'is this drug safe?' alongside PRR/ROR and 'label-vs-real-world AE'.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete integrations — 'FDA labels, FAERS adverse event reports, PRR/ROR disproportionality, pharmacogenomic biomarkers, clinical trial data, and published literature' — matching the multiple-specific-actions anchor.	3 / 3
Completeness	Explicitly states both what it does ('End-to-end drug safety review integrating...') and when to use it ('Use for regulatory drug safety reviews, comprehensive pharmacovigilance reports...'), satisfying the what-AND-when anchor.	3 / 3
Trigger Term Quality	Natural terms like 'drug safety review', 'pharmacovigilance reports', and 'clinical decision support' are present, but specialist jargon ('PRR/ROR disproportionality', 'label-vs-real-world AE comparison') is less likely to be phrased that way by a typical user, leaving some common variations uncovered.	2 / 3
Distinctiveness Conflict Risk	The description carves a clear drug-safety-integration niche and even names adjacent skills ('tooluniverse-adverse-event-detection', 'tooluniverse-pharmacovigilance'), making wrong-skill triggering unlikely.	3 / 3
	Total	11 / 12 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-clinical-data-integration/SKILL.md
Commit: 089eb8e

Reviewed: about 6 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.