tooluniverse-protein-structure-prediction

Protein 3D structure prediction from sequence — ESMFold de novo prediction, AlphaFold database retrieval, experimental structures from RCSB, ProtVar variant impact assessment, ProtParam sequence properties. Use for structure prediction when no experimental structure exists, fold-confidence scoring, and structure-guided variant interpretation.

Quality

88%

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

77%

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

A thorough, well-sequenced workflow with highly actionable tool guidance and clear checkpoints. Its main weaknesses are token redundancy from repeated parameter tables and a monolithic structure with no progressive disclosure via bundle files.

Suggestions

Dedupe tool parameter documentation: keep parameters in the per-phase sections OR in the Tool Parameter Reference table, not both, to remove ~30 redundant lines.

Move the Tool Parameter Reference, Databases Integrated, and Limitations tables into a references file (e.g. references/tools.md) and link to it from SKILL.md so detail is one level deep.

Trim the generic 'LOOK UP, DON'T GUESS' and 'COMPUTE, DON'T DESCRIBE' sections to a single line each, or fold them into the KEY PRINCIPLES list.

Dimension	Reasoning	Score
Conciseness	Mostly efficient operational content, but tool parameters are repeated across each phase and again in the Input Parameters and Tool Parameter Reference tables, and the generic 'LOOK UP, DON'T GUESS' / 'COMPUTE, DON'T DESCRIBE' sections pad the token budget. Not level 3 because of this redundancy; not level 1 because the bulk is concrete, domain-specific guidance rather than concepts Claude already knows.	2 / 3
Actionability	Concrete tool names, exact parameters, example values (e.g. 'P04637 R175H'), return fields, decision logic, fallback strategies, and quality minimums give copy-ready guidance for a tool-call skill. The scoring note permits absence of raw code when instruction guidance is this actionable.	3 / 3
Workflow Clarity	A clearly sequenced Phase 0–6 workflow with a diagram, per-phase objectives/tools/steps, explicit decision logic, fallback strategies, and quality minimums that act as validation checkpoints. Not level 2 because checkpoints are explicit rather than implicit, and the destructive/batch cap does not apply to this read-only prediction workflow.	3 / 3
Progressive Disclosure	Internally well-organized into labeled phases, but it is a single ~334-line monolithic file with no bundle files (references/scripts/assets absent) and no one-level-deep references to offload detail. Not level 3 because nothing is split out for a long document; not level 1 because it is structured, not a wall of text.	2 / 3
	Total	10 / 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.

A strong, third-person description that names concrete capabilities and pairs them with explicit use-when triggers in a distinctive niche. It cleanly answers both what the skill does and when to invoke it.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete actions — 'ESMFold de novo prediction, AlphaFold database retrieval, experimental structures from RCSB, ProtVar variant impact assessment, ProtParam sequence properties' — matching the score-3 anchor that lists several specific actions.	3 / 3
Completeness	Explicitly answers both 'what' (named capabilities/tools) and 'when' via an explicit 'Use for ...' clause, satisfying the score-3 anchor; it is not the level below because the trigger guidance is explicit rather than implied.	3 / 3
Trigger Term Quality	'Use for structure prediction when no experimental structure exists, fold-confidence scoring, and structure-guided variant interpretation' provides natural domain triggers a user would actually say, not just internal jargon.	3 / 3
Distinctiveness Conflict Risk	Named tools and a tightly scoped protein-structure niche give it clear, distinct triggers unlikely to fire for unrelated skills; not level 2 because it is far more specific than a generic 'works with files' description.	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-protein-structure-prediction/SKILL.md
Commit: 089eb8e

Reviewed: about 7 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.