tooluniverse-polygenic-risk-score

Build and interpret polygenic risk scores (PRS) for complex diseases using GWAS summary statistics. Covers PRS construction (clumping/thresholding, PRS-CS), validation in independent cohorts, ancestry-aware adjustment, and clinical interpretation (population-relative risk, not absolute prediction). Use for PRS-based risk stratification.

Quality

75%

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

Fix and improve this skill with Tessl

tessl review fix ./plugin/skills/tooluniverse-polygenic-risk-score/SKILL.md

Quality

Content

50%

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

The body provides rich, domain-specific operational guidance with real tools and concrete thresholds, but is padded with a Key Concepts section explaining genetics fundamentals Claude already knows, uses invented wrapper functions in the workflow code rather than the actual described tools, and lacks validation checkpoints in the workflow. With no bundle files, the large monolithic file should offload reference material into separate files.

Suggestions

Cut or move the "Key Concepts" section (GWAS, odds ratios, LD, population stratification basics) — Claude already knows these; keep only the PRS-specific operational guidance to improve conciseness.

Replace the invented wrapper functions (build_polygenic_risk_score, calculate_personal_prs, interpret_prs_percentile) with actual calls to the described ToolUniverse tools (gwas_search_associations, gnomad_search_variants, run_ldsc_heritability) so the code is executable.

Add explicit validation checkpoints to the workflow (e.g. confirm QC filters were applied and sufficient genome-wide-significant SNPs remain before score calculation, and verify population mean/std before percentile interpretation).

Split detailed reference material (Data Sources tool reference, Key Concepts, Applications) into bundle files under references/ and link to them one level deep to reduce the monolithic body.

Dimension	Reasoning	Score
Conciseness	The body is ~280 lines and the "Key Concepts" section re-explains GWAS, odds ratios, beta coefficients, LD/clumping, and population stratification — concepts Claude already knows — so it is mostly efficient but includes unnecessary explanation that could be tightened, matching the level-2 anchor rather than the lean level-3.	2 / 3
Actionability	Real tools with parameters (e.g. gwas_search_associations with disease_trait/size) and concrete QC thresholds (MAF < 0.01, HWE p < 1e-6) are actionable, but the headline workflow code calls invented wrappers (build_polygenic_risk_score, calculate_personal_prs, interpret_prs_percentile) that are not among the described ToolUniverse tools, fitting the level-2 "some concrete guidance but incomplete / pseudocode" anchor.	2 / 3
Workflow Clarity	The 6-step Workflow (Trait Selection → ... → Risk Interpretation) is clearly sequenced, but there are no explicit validation checkpoints or feedback loops for a batch/data operation, which per the guidelines caps workflow clarity at 2.	2 / 3
Progressive Disclosure	Sections are well-organized (not a disorganized wall of text), but the skill is a single ~280-line monolithic file with no bundle files (references/scripts/assets are absent) and content such as Key Concepts, Data Sources, and Applications that could be split remains inline, matching the level-2 "content that should be separate is inline" anchor.	2 / 3
	Total	8 / 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 concise, specific, and uses third-person voice with concrete construction/validation/interpretation actions. It explicitly states both what the skill does and when to use it via an explicit trigger clause. Voice is correctly third person ("Build and interpret") with no first/second-person penalty.

Dimension	Reasoning	Score
Specificity	Lists multiple specific concrete actions — "PRS construction (clumping/thresholding, PRS-CS), validation in independent cohorts, ancestry-aware adjustment, and clinical interpretation" — rather than vague language, matching the level-3 anchor.	3 / 3
Completeness	Clearly answers both what ("Build and interpret polygenic risk scores...") and when via the explicit "Use for PRS-based risk stratification" trigger, satisfying the level-3 anchor for both what AND when.	3 / 3
Trigger Term Quality	Includes natural terms users would say — "polygenic risk scores (PRS)", "GWAS", "PRS-based risk stratification" — giving good coverage of the domain's common vocabulary, matching the level-3 anchor.	3 / 3
Distinctiveness Conflict Risk	The polygenic risk score niche with PRS/GWAS triggers is clearly distinguishable and unlikely to fire for unrelated skills, matching the level-3 "clear niche with distinct triggers" anchor.	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-polygenic-risk-score/SKILL.md
Commit: 089eb8e

Reviewed: about 9 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.