clinical-decision-support

Prepare and validate research-only clinical decision-support evaluation, evidence-profile, cohort, survival, biomarker/model, privacy, and governance artifacts. Use for aggregate or synthetic research documentation and traceability—not patient care or live clinical operation.

Quality

83%

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

78%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.

The body is well-structured and genuinely actionable, with strong progressive disclosure that keeps the overview lean while routing detail to verified reference files. The main gaps are minor: restated safety language, --help-only run examples, and no explicit error-recovery loop in verification.

Suggestions

Consolidate the repeated safety-boundary statements into the Hard Safety Boundary section and reference it once from downstream sections to trim redundant tokens.

Replace at least one "--help" line in Run Locally with a full end-to-end example (e.g., `python3 scripts/cohort_table_generator.py assets/aggregate_cohort_table_template.json -o out/`) so users see the actual invocation pattern.

Add an explicit feedback loop to the Verification section (e.g., "If tests or AST parse fail, fix the script and re-run until both pass before proceeding").

Dimension	Reasoning	Score
Conciseness	The body is dense and assumes Claude's competence—no padding with basic concept explanations—and uses compact tables and bullets; not a 5 because the safety boundary is restated across multiple sections (Hard Safety Boundary, Data Gate, Required Artifact Header, per-topic "Do not..." lines), adding tokens that could be consolidated.	4 / 5
Actionability	Provides a concrete need→asset→script selection table, an executable Verification block ("python3 -m unittest discover...", "python3 -c \"import ast...\""), and a Data Gate checklist; not a 5 because the Run Locally section shows only "--help" invocations rather than a full example of running a script with real input to produce output.	4 / 5
Workflow Clarity	A clear four-step Workflow (Frame → Select → Run → Human Review) with a前置 Data Gate validation checkpoint, a Required Artifact Header checklist, and a Verification section; not a 5 because the Verification block lacks an explicit validate→fix→retry feedback loop for the batch/destructive-adjacent operations.	4 / 5
Progressive Disclosure	SKILL.md acts as an overview with each topic ending in a well-signaled one-level-deep "See `references/X.md`" pointer, a terminal Reference Map for navigation, and a selection table; all referenced references/scripts/assets files resolve to real bundle files.	5 / 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.

The description is strong: it states concrete actions over a comprehensive artifact set, gives an explicit "Use for" trigger clause, and carves out a distinctive research-only niche with a negative boundary. Minor gaps in action-verb variety and synonym coverage keep specificity and trigger_term_quality just below 5.

Dimension	Reasoning	Score
Specificity	Phrases "Prepare and validate" give two concrete actions applied to a comprehensive list of seven artifact types ("evaluation, evidence-profile, cohort, survival, biomarker/model, privacy, and governance artifacts"), matching the several-specific-actions anchor; not a 5 because the verbs are fewer and somewhat generic compared to the 5-anchor's four concrete verbs.	4 / 5
Completeness	Explicitly answers both what ("Prepare and validate ... artifacts") and when ("Use for aggregate or synthetic research documentation and traceability—not patient care or live clinical operation") with concrete trigger phrases, mirroring the 5-anchor structure; the negative boundary further sharpens the when.	5 / 5
Trigger Term Quality	Includes natural domain terms a user would say ("clinical decision-support", "evidence-profile", "cohort", "survival", "biomarker/model", "aggregate or synthetic research documentation and traceability") with good coverage; not a 5 because it lacks synonyms and file-extension variants.	4 / 5
Distinctiveness Conflict Risk	Occupies a clear, specialized niche (research-only CDS evaluation) with distinct triggers and an explicit exclusion ("not patient care or live clinical operation") that minimizes conflict with other skills.	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: K-Dense-AI/scientific-agent-skills
Path: skills/clinical-decision-support/SKILL.md
Commit: b2a92ba

Reviewed: 3 days 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.