Name: tooluniverse-protein-structure-prediction
Rating: 64.8 (1 reviews)
Author: mims-harvard

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

76%

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

Fix and improve this skill with Tessl

tessl review fix ./plugins/tooluniverse/skills/tooluniverse-protein-structure-prediction/SKILL.md

The canonical home for this skill is tooluniverse-protein-structure-prediction in mims-harvard/ToolUniverse

Quality

Content

63%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, actionable workflow document with clear phases and tool guidance. It loses points for verbosity from generic/duplicated sections and for keeping all detail inline rather than offloading reference material to bundle files.

Suggestions

Remove the generic 'LOOK UP, DON'T GUESS' and 'COMPUTE, DON'T DESCRIBE' sections or replace them with protein-structure-specific guidance; they pad tokens without adding skill-specific value.

Move the Tool Parameter Reference, Databases Integrated, and Fallback Strategies tables into a references/ file (e.g. TOOL_REFERENCE.md) and link to them from the body to improve progressive disclosure.

Consolidate the Quality Minimums into an explicit validate-then-report gate at the end of Phase 6 so the workflow's checkpoints are unmistakable.

Dimension	Reasoning	Score
Conciseness	Mostly efficient but includes generic padding ('LOOK UP, DON'T GUESS', 'COMPUTE, DON'T DESCRIBE') unrelated to this skill and redundant tables (Databases Integrated, Fallback Strategies) that duplicate the per-phase content.	3 / 5
Actionability	Gives concrete tool names with specific parameters and format examples (e.g. '"P04637 R175H"', qualifier/uniprot_id); code absence is acceptable for a tool-orchestration skill, though no complete example call is shown.	4 / 5
Workflow Clarity	A clearly sequenced six-phase workflow with decision logic, fallbacks, and a Quality Minimums checklist; checkpoints are present but somewhat distributed rather than consolidated as explicit validate-then-proceed gates.	4 / 5
Progressive Disclosure	At ~340 lines with no bundle files, the SKILL.md is a monolith inlining a tool-parameter reference, databases table, fallbacks, and six detailed phases that could be split into reference files; no external references are signaled.	3 / 5
	Total	14 / 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, specific description that clearly states capabilities and explicit use conditions. Main weakness is jargon-heavy trigger language and an undrawn boundary against the related retrieval skill.

Suggestions

Add plain-language trigger synonyms (e.g., 'predict protein structure', '3D model of a protein') alongside the tool names so non-expert phrasing matches.

Add a brief carve-out from retrieval-only tasks in the description itself, e.g. 'Not for retrieving a known PDB/AlphaFold structure without prediction.'

Dimension	Reasoning	Score
Specificity	Lists multiple concrete actions (ESMFold de novo prediction, AlphaFold retrieval, RCSB experimental structures, ProtVar variant impact, ProtParam properties) with comprehensive coverage matching the anchor 5 example.	5 / 5
Completeness	Explicitly answers both what (the enumerated actions) and when via 'Use for structure prediction when no experimental structure exists, fold-confidence scoring, and structure-guided variant interpretation.'	5 / 5
Trigger Term Quality	Includes natural terms like 'structure prediction' and 'fold-confidence scoring' but leans heavily on tool jargon (ESMFold/AlphaFold/RCSB/ProtVar/ProtParam) and omits common synonyms a user would say such as 'predict protein structure'.	4 / 5
Distinctiveness Conflict Risk	Occupies a clear protein-structure-prediction niche, but the description does not distinguish it from the sibling tooluniverse-protein-structure-retrieval skill (noted only in the body), leaving minor overlap risk.	4 / 5
	Total	18 / 20 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: plugins/tooluniverse/skills/tooluniverse-protein-structure-prediction/SKILL.md
Commit: 1b02920

Reviewed: 1 day 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.