tooluniverse-variant-functional-annotation

Functional annotation of protein variants — ProtVar structural/functional context, ClinVar clinical classifications, gnomAD population frequencies, CADD deleteriousness, ClinGen gene-disease validity, plus FAVOR one-call comprehensive GRCh38 annotation. Use for variant annotation pipelines, missense effect prediction, and protein-level variant interpretation with functional context.

Quality

85%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

Quality

Content

70%

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

The body is well-structured with a clear multi-phase workflow, fallback chains, and a concrete report template, scoring high on workflow clarity. It loses points for re-teaching known genetics concepts (conciseness), lacking executable code examples (actionability), and keeping all detail inline with no bundle references (progressive disclosure).

Suggestions

Trim the Pathogenicity Reasoning Framework's explanatory prose about basic genetics concepts (conservation, population frequency logic) down to decision rules and thresholds Claude can apply, preserving the actionable thresholds without re-teaching the underlying biology.

Add at least one executable code or command example (e.g., a Python snippet invoking one of the ToolUniverse tools with sample parameters) to lift actionability from procedural prose to copy-paste-ready guidance.

Move the per-tool parameter reference, fallback chains, and the full report template into separate reference files (e.g., TOOLS.md, REPORT-TEMPLATE.md) referenced one level deep from SKILL.md to improve progressive disclosure.

Dimension	Reasoning	Score
Conciseness	Largely efficient and assumes Claude's domain competence, but the Pathogenicity Reasoning Framework section restates well-known genetics concepts (e.g., that conserved residues tolerate mutation poorly, that high population AF argues against severe Mendelian disease) that Claude already knows, adding length without proportional value.	2 / 3
Actionability	Tool names, parameter signatures, and thresholds (CADD >= 30, gnomAD AF < 0.001) are concrete and specific, but there are no executable code or command examples; guidance is procedural prose rather than copy-paste-ready instructions, which the rubric treats as incomplete actionability.	2 / 3
Workflow Clarity	The workflow is clearly sequenced into Phases 0-5 plus Synthesis with an ASCII diagram, includes fallback chains for each phase, and embeds decision points (e.g., annotator selection by variant type, ClinGen gating ClinVar interpretation) that act as validation/recovery checkpoints.	3 / 3
Progressive Disclosure	No bundle files (references/scripts/assets) exist, so the skill is monolithic; while it is well-sectioned, large reference-style content (full tool parameter specs, the report template, fallback chains) remains inline in SKILL.md rather than split into one-level-deep referenced files.	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 (both what and when), and explicitly differentiated from a sibling skill. It uses appropriate third-person voice and avoids fluff. This is a strong description that aligns with the rubric's good examples.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete actions: 'Functional annotation of protein variants', retrieving ProtVar/ClinVar/gnomAD/CADD/ClinGen/FAVOR annotations, supporting missense effect prediction and protein-level interpretation.	3 / 3
Completeness	Clearly states what it does (functional annotation combining six evidence sources) and when to use it via an explicit 'Use for...' clause covering variant annotation pipelines, missense effect prediction, and protein-level variant interpretation.	3 / 3
Trigger Term Quality	Includes natural user-facing terms users would say: 'variant annotation pipelines', 'missense effect prediction', 'population frequencies', and tool names (ClinVar, gnomAD, CADD) that match real query phrasing.	3 / 3
Distinctiveness Conflict Risk	It carves out a clear protein-level functional-evidence niche and explicitly distinguishes itself from `tooluniverse-variant-interpretation`, making it unlikely to trigger the wrong skill.	3 / 3
	Total	12 / 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-variant-functional-annotation/SKILL.md
Commit: 089eb8e

Reviewed: about 8 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.