tooluniverse-statistical-modeling

Statistical modeling — linear/logistic/ordinal/Poisson regression, ANOVA, Kruskal-Wallis, chi-square, Mann-Whitney, Cox survival, spline fits (R `ns()`), odds ratios, Cohen's d, F-statistic, p-value computation. Specializes in clinical-trial AE analysis (SDTM DM/AE), severity ordinal regression, and per-feature stat workflows.

Quality

70%

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-statistical-modeling/SKILL.md

Quality

Content

67%

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

The content is exceptionally actionable with concrete executable scripts and invocations, and it has a sensible RULE-ZERO gating plus bundle references. It is held back by verbosity and repetition (re-stating AE/spline guidance, explaining concepts Claude knows) and by a monolithic SKILL.md with missing validation checkpoints and a few referenced files absent from the bundle.

Suggestions

Move the Model Selection Decision Tree, the evidence-grading/interpretation tables, and the repeated AE-severity and spline conventions into reference files, keeping SKILL.md a lean overview with one-level-deep pointers.

Add explicit validation checkpoints to the workflow (e.g. 'run diagnostics; if assumption violated, switch model; only then report') so multi-step analyses have a clear validate-then-proceed loop.

Reconcile the File Structure listing with the actual bundle so all referenced files (anova_and_tests.md, common_patterns_summary.md, QUICK_START.md, EXAMPLES.md, TOOLS_REFERENCE.md) exist, or remove the dangling references.

Dimension	Reasoning	Score
Conciseness	The body is information-dense and mostly earns its tokens, but it includes content Claude already knows (a full Model Selection Decision Tree, library-vs-library tables, generic interpretation guidance like 'OR = 1.0 means no association') and repeats the AE-severity / spline guidance in multiple places (PRIMARY SCRIPTS, CRITICAL section, Analysis conventions). It is efficient in places but could be tightened; not a level-3 'every token earns its place' lean file.	2 / 3
Actionability	It provides fully executable, copy-paste-ready bash invocations for every bundled script with concrete flags, plus runnable Python snippets (pearsonr, statsmodels formulas, cohort prep). This matches the 'fully executable code/commands; copy-paste ready' anchor.	3 / 3
Workflow Clarity	There is a clear RULE ZERO pre-computed-results-first gate and a phased Workflow (Phase 0-3), but validation/checkpoint steps are light: Phase 2 lists diagnostics as a reference pointer rather than an explicit validate-then-proceed loop, and the bundled-script path has no explicit failure/retry checkpoint. Per the judging notes, missing validation checkpoints in multi-step computational workflows caps clarity at 2.	2 / 3
Progressive Disclosure	There is real bundle structure (references/ and scripts/ with well-signaled one-level-deep pointers like 'See references/logistic_regression.md'), but the SKILL.md itself is a monolithic ~660-line wall that keeps large amounts of inline detail (full decision trees, interpretation tables, repeated conventions) that belong in references. Several referenced files (anova_and_tests.md, common_patterns_summary.md, QUICK_START.md, EXAMPLES.md, TOOLS_REFERENCE.md) are cited but not present in the bundle.	2 / 3
	Total	9 / 12 Passed

Description

72%

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 highly specific and distinctive, naming concrete statistical methods and a clear biomedical niche in third person. Its main weakness is the absence of an explicit 'Use when...' trigger clause, which caps completeness, and the trigger terms lean technical rather than natural.

Suggestions

Add an explicit 'Use when...' clause naming natural user phrasings (e.g. 'Use when computing odds ratios, p-values, or fitting regression on clinical-trial AE/SDTM data').

Soften or supplement the technical jargon ('R `ns()`', 'SDTM DM/AE', 'per-feature stat workflows') with plain-language keywords a user would actually say.

Dimension	Reasoning	Score
Specificity	Lists many concrete actions and methods (linear/logistic/ordinal/Poisson regression, ANOVA, Cox survival, spline fits, odds ratios, Cohen's d, F-statistic, p-value computation) plus a concrete specialization (clinical-trial AE analysis on SDTM DM/AE). This matches the 'multiple specific concrete actions' anchor.	3 / 3
Completeness	It thoroughly answers 'what does this do' with capabilities and a specialization, but there is no 'Use when...' clause or explicit trigger guidance for when Claude should invoke it. Per the judging guidelines, a missing explicit trigger clause caps completeness at 2.	2 / 3
Trigger Term Quality	It includes domain terms a user might say (odds ratio, p-value, ANOVA, survival), but the phrasing is a dense technical enumeration (e.g. 'R `ns()`', 'SDTM DM/AE', 'per-feature stat workflows') rather than natural user-facing keywords like 'analyze clinical trial data' or 'compute a p-value'. Coverage of natural phrasings is partial.	2 / 3
Distinctiveness Conflict Risk	The clinical-trial AE / SDTM / ordinal-severity niche and the R natural-spline specifics are a clear, narrow domain unlikely to conflict with other skills, and third-person voice is used throughout.	3 / 3
	Total	10 / 12 Passed

Validation

93%

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 — 15 / 16 Passed

Validation for skill structure

Criteria	Description	Result
skill_md_line_count	SKILL.md is long (660 lines); consider splitting into references/ and linking	Warning

	Total	15 / 16 Passed

Repository: mims-harvard/ToolUniverse
Path: plugins/tooluniverse/skills/tooluniverse-statistical-modeling/SKILL.md
Commit: 089eb8e

Reviewed: about 7 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.