CtrlK
BlogDocsLog inGet started
Tessl Logo

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.

67

Quality

80%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

SecuritybySnyk

Low

Low-risk findings worth noting

Fix and improve this skill with Tessl

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

The canonical home for this skill is tooluniverse-adverse-event-detection in mims-harvard/ToolUniverse

SKILL.md
Quality
Evals
Security

Quality

Content

71%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, highly actionable pipeline skill with concrete tools, thresholds, edge cases, and a phased workflow. Its main defect is that all four referenced detail files (PHASE_DETAILS.md, REPORT_TEMPLATE.md, TOOL_REFERENCE.md, QUICK_START.md) are missing from the bundle, so the progressive-disclosure structure promises content it does not deliver; secondary issues are minor duplication between the workflow diagram and phase summaries.

Suggestions

Add the four referenced bundle files (PHASE_DETAILS.md, REPORT_TEMPLATE.md, TOOL_REFERENCE.md, QUICK_START.md) — or stop referencing them — since none currently exist in the skill directory.

Inline the essential content that the workflow depends on (the signal strength classification table and the Safety Signal Score rubric) in SKILL.md so the pipeline is executable even without the missing PHASE_DETAILS.md.

Trim the duplicated 10-phase ASCII workflow diagram, or the Phase Summaries, since both convey the same sequence.

DimensionReasoningScore

Conciseness

The body is dense with domain-specific value (tool names per phase, signal thresholds like "PRR >= 2.0 AND lower CI > 1.0 AND N >= 3", edge cases) and mostly assumes Claude's competence. Minor trimmable padding: the 10-phase ASCII workflow diagram largely duplicates the Phase Summaries, and the Naranjo algorithm walkthrough re-explains a framework Claude largely knows. Not anchor 5 ('every token earns its place') because of that duplication, and clearly not anchor 3's 'unnecessary explanation' territory.

4 / 5

Actionability

Concrete, executable guidance throughout: exact tool names per phase (e.g. "FAERS_calculate_disproportionality", "OpenTargets_get_drug_chembId_by_generic_name"), explicit signal criteria and strength cutoffs, and a point-weighted scoring rubric ("FAERS signal strength (0-35), serious AEs (0-30), FDA label warnings (0-25), literature evidence (0-10)"). Not anchor 5 because no inline executable examples or tool-call snippets appear — examples and output templates are entirely deferred to reference files.

4 / 5

Workflow Clarity

A clear 10-phase sequence with per-phase outputs, explicit signal criteria acting as checkpoints, error-handling guidance ("{error: {code: \"NOT_FOUND\"}} is normal when a section does not exist"), an Edge Cases section with fallbacks ("No FAERS reports: Skip Phases 1-2"), and a completeness checklist in Phase 9. Not anchor 5 because validation between phases is implicit (criteria stated but no explicit verify-then-branch loop like the validate/fix/re-validate example).

4 / 5

Progressive Disclosure

References are clearly signaled and one level deep ("Reference files (in this directory): PHASE_DETAILS.md, REPORT_TEMPLATE.md, TOOL_REFERENCE.md, QUICK_START.md" plus inline "See PHASE_DETAILS.md for scoring rubric"), and critical details are correctly split out. However, none of the four referenced files exist in the bundle — no references/, scripts/, or assets/ directories are present — so the disclosure chain is broken and the promised detail (signal classification table, scoring rubric, report template) is actually inaccessible. This lands at anchor 3 ('structure present but not fully delivering') rather than 4.

3 / 5

Total

15

/

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 description that names concrete capabilities (PRR/ROR/IC disproportionality statistics, 0-100 safety signal scores, evidence grading) and pairs them with an explicit, naturally phrased 'Use for' trigger clause. Its only weaknesses are a few missing user synonyms and no in-description differentiation from a general pharmacovigilance skill.

DimensionReasoningScore

Specificity

Multiple specific concrete actions are named: "Detect and analyze adverse drug event signals using FDA FAERS reports, drug labels, and disproportionality statistics (PRR, ROR, IC)" and "Generates quantitative safety signal scores (0-100) with evidence grading". Specific data sources, statistics, and output format are all explicit; not the vague anchor below ("Processes PDF files"-level generality) because capabilities are enumerated with concrete parameters.

5 / 5

Completeness

Explicitly answers both questions: a concrete 'what' (detect/analyze AE signals with named statistics; generate 0-100 scores with evidence grading) and an explicit 'when' clause ("Use for post-market surveillance, pharmacovigilance, drug safety assessment, regulatory submissions, and detecting rare AE signals not visible in clinical trials") with concrete trigger phrases, matching the top anchor.

5 / 5

Trigger Term Quality

Natural user-facing phrases present: "post-market surveillance", "pharmacovigilance", "drug safety assessment", "regulatory submissions", "FAERS", "adverse drug event signals". Falls between anchors 4 and 5: good coverage but misses common synonyms a user would say, e.g. "side effects", "adverse reactions", "drug safety profile".

4 / 5

Distinctiveness Conflict Risk

The FAERS/disproportionality/safety-score framing carves a clear niche with minimal conflict risk against generic skills. However, it does not self-distinguish from a sibling pharmacovigilance skill — the differentiation (signal quantification vs. general pharmacovigilance) lives only in the body, so minor overlap risk remains.

4 / 5

Total

18

/

20

Passed

Validation

100%

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
Reviewed

Table of Contents

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.