tooluniverse-protein-lof-mechanism

Propose the mechanism by which a missense variant causes loss-of-function (LoF), synthesizing evidence from 5 independent layers: AlphaMissense pathogenicity, AlphaFold structural context, ESMC sequence likelihood, SAE feature disruption, and DynaMut2 stability ΔΔG. Distinguishes 'structural stability LoF' (mis-folding) from 'direct functional disruption' (catalytic / binding / PTM site damage). Use for coding missense variants where you need a mechanistic causal model, not just a pathogenicity score.

Quality

86%

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

Quality

Content

85%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 highly actionable, well-sequenced workflow with executable code and genuine validation gating. The main weakness is progressive disclosure: everything lives in one long inline file with no separation of the optional/advanced material into reference files.

Suggestions

Move the optional ThermoMPNN section (local install, SaaS platform list, when-it-matters guidance) into a reference file like THERMOMPNN.md and link to it from a short pointer, keeping the main workflow lean.

Consider extracting the 6-category synthesis decision table and/or the limitations list into a separate reference so the core 7-step workflow reads as an overview.

Tighten the commercial SaaS platform bullet list to the 1-2 most useful options or move it entirely to the reference file to improve token efficiency.

Dimension	Reasoning	Score
Conciseness	Mostly efficient with executable code, threshold tables, and a decision matrix that earn their tokens, but the ~45-line optional ThermoMPNN section (5 commercial SaaS links plus citation) and a long limitations list add length that could be trimmed.	4 / 5
Actionability	Fully executable, copy-paste-ready tool calls with real arguments and expected outputs throughout (UniProt_search, AlphaMissense_get_variant_score, DynaMut2_predict_stability with pdb_id/chain/mutation), plus a signal-pattern decision table mapping to concrete rescue strategies.	5 / 5
Workflow Clarity	Clear Step 0-7 sequence with explicit validation checkpoints (reference-residue assert, pLDDT and AlphaMissense gating) and an honest evidence-grading step with high/medium/low criteria acting as a feedback loop for low-confidence cases.	5 / 5
Progressive Disclosure	Well-organized with clear section headers, but the ~290-line skill is monolithic with no bundle files or external references, and advanced material (the 5-platform SaaS list, ThermoMPNN install details, limitations) is inlined rather than split into reference files.	3 / 5
	Total	17 / 20 Passed

Description

87%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 both the capability and the trigger conditions in third person, with named tools and a sharp niche boundary. The only soft spots are trigger-term synonym coverage and the action list being framed around one main verb.

Suggestions

Add common user-facing synonyms such as 'amino acid substitution' or 'point mutation' alongside 'missense variant' to broaden natural trigger coverage.

Lead with a short verb-list of the concrete actions (synthesize signals, classify mechanism, grade confidence) before the prose to make the capability list read as comprehensive.

Dimension	Reasoning	Score
Specificity	Lists several concrete actions ('Propose the mechanism', 'synthesizing evidence from 5 independent layers', 'Distinguishes structural stability LoF from direct functional disruption') with the 5 named signals, but centers on one main verb with sub-actions rather than a comprehensive list of discrete operations.	4 / 5
Completeness	Explicitly answers both 'what' (propose the LoF mechanism by synthesizing 5 evidence layers) and 'when' ('Use for coding missense variants where you need a mechanistic causal model, not just a pathogenicity score') with concrete trigger phrases, matching the top anchor.	5 / 5
Trigger Term Quality	Includes domain-natural terms a variant-interpretation user would say ('missense variant', 'loss-of-function (LoF)', 'coding missense variants', 'pathogenicity score', 'mechanistic causal model'), but misses common synonyms like 'amino acid substitution' or 'point mutation'.	4 / 5
Distinctiveness Conflict Risk	Occupies a clear niche (missense-variant LoF mechanism synthesis across 5 named tools) and sharpens its boundary ('not just a pathogenicity score'), with the body carving out non-coding/ACMG/cancer skills, giving minimal conflict risk.	5 / 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: plugin/skills/tooluniverse-protein-lof-mechanism/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.