CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

decision-curve-analysis

Use when evaluating the clinical utility of a binary prediction model from a single clinical CSV file by fitting a logistic decision-curve model, plotting decision and clinical-impact curves, and exporting summary outputs. NOT for: survival calibration, ROC-only discrimination analysis, nomogram construction, or time-to-event outcomes.

Quality

83%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Passed

No findings from the security scan

SKILL.md
Quality
Evals
Security

Quality

Content

67%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, highly actionable body: complete CLI reference, concrete input/output specs, a validated workflow, and clearly signaled reference files. Main weaknesses are redundant sections that waste tokens and a Testing section that references bundled test files which are missing from the actual bundle.

Suggestions

Remove the redundant "Basic Usage" example (identical to the Usage block) and merge the "Input Validation" exclusions into "When Not to Use" to cut duplicated content; drop the dev-facing "Implementation Checklist" and the version/date footer from the body.

Ship the referenced `tests/` files (tests/data/dca_data.csv, tests/run_smoke_test.R/.sh) or update the Testing and "With Bundled Test Data" sections to reference files that exist, so the smoke-test verification step is runnable.

Add "DCA" and "net benefit" as trigger terms in the frontmatter description, since these are the phrases users most commonly use when requesting decision curve analysis.

DimensionReasoningScore

Conciseness

Mostly efficient (argument/error tables, sample CSV, no concept tutoring), but several sections could be trimmed: the "Basic Usage" example repeats the Usage block verbatim, the "Input Validation" section re-enumerates the same exclusions as "When Not to Use", the dev-facing "Implementation Checklist" is padding, and the "Last updated: 2026-04-27 | Version: 1.1.0" footer is time-sensitive info outside a deprecation section. This fits 'mostly efficient but includes some unnecessary explanation or could be tightened' rather than 4, where trims would be minor.

3 / 5

Actionability

Fully executable copy-paste Rscript commands, a complete argument table with types/defaults, error codes mapped to causes and solutions, and a sample CSV format. Not 5 because the Testing section and the "With Bundled Test Data" example reference `tests/data/dca_data.csv`, `tests/run_smoke_test.R`, and `tests/run_smoke_test.sh`, which are not present in the bundle, so those commands are not runnable as written.

4 / 5

Workflow Clarity

The 4-step Workflow has explicit validation up front ("Validate Input", "Reject cohorts with too few rows or too few events/non-events") and an error-recovery loop (error table → "IF error persists, READ: references/troubleshooting.md"). Not 5 because the verification path (smoke test) depends on test files that do not exist in the bundle, so the end-to-end validate-fix-verify loop cannot actually be executed as documented.

4 / 5

Progressive Disclosure

Good structure with a dedicated "When to Read External Files" table signaling one-level-deep references, and the referenced files (`references/algorithm.md`, `references/cli-guide.md`, `references/troubleshooting.md`, `scripts/main.R`) all exist in the bundle. Not 5 because the body also points to a `tests/` directory that is absent from the bundle, and some duplicated usage material (Usage vs. Examples, exclusion lists in two sections) sits inline instead of being consolidated or offloaded.

4 / 5

Total

15

/

20

Passed

Description

92%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 description: concrete what, explicit when, a well-defined single-CSV input scope, and NOT-for boundaries that sharply reduce conflict risk. The only gap is missing high-frequency synonyms ("DCA", "net benefit") that users commonly say when requesting this analysis.

DimensionReasoningScore

Specificity

Quotes multiple concrete actions with a comprehensive pipeline: "fitting a logistic decision-curve model, plotting decision and clinical-impact curves, and exporting summary outputs", plus a concrete input spec ("from a single clinical CSV file"). This matches the anchor 'Lists multiple specific concrete actions; comprehensive coverage' — not 4, since coverage spans input, modeling, plotting, and export with no gaps.

5 / 5

Completeness

Explicitly answers both: what ("fitting a logistic decision-curve model, plotting decision and clinical-impact curves, and exporting summary outputs") and when ("Use when evaluating the clinical utility of a binary prediction model from a single clinical CSV file"), with concrete trigger phrasing plus an explicit NOT-for exclusion list. Matches the anchor-5 example structure exactly.

5 / 5

Trigger Term Quality

Good natural-term coverage: "clinical utility", "binary prediction model", "decision-curve", "clinical-impact curves", "clinical CSV". Not 5 because common user phrasings are missing: the abbreviation "DCA" and "net benefit" never appear, which are the most frequent terms users say for this task; not 3 since most relevant keywords and synonyms are present.

4 / 5

Distinctiveness Conflict Risk

A clear niche (decision curve analysis of binary prediction models) with explicit boundary-setting: "NOT for: survival calibration, ROC-only discrimination analysis, nomogram construction, or time-to-event outcomes". This minimally conflicts with neighboring survival/ROC/nomogram skills; not 4 because the exclusions actively disambiguate the closest related skills.

5 / 5

Total

19

/

20

Passed

Validation

93%

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 15 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

frontmatter_unknown_keys

Unknown frontmatter key(s) found; consider removing or moving to metadata

Warning

Total

15

/

16

Passed

Repository

aipoch/medical-research-skills

Path

awesome-med-research-skills/Data Analysis/decision-curve-analysis/SKILL.md

Reviewed

Table of Contents

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.