scikit-survival
Comprehensive toolkit for survival analysis and time-to-event modeling in Python using scikit-survival. Use this skill when working with censored survival data, performing time-to-event analysis, fitting Cox models, Random Survival Forests, Gradient Boosting models, or Survival SVMs, evaluating survival predictions with concordance index or Brier score, handling competing risks, or implementing any survival analysis workflow with the scikit-survival library.
89
78%
Does it follow best practices?
Impact
96%
1.24xAverage score across 6 eval scenarios
Passed
No known issues
Optimize this skill with Tessl
npx tessl skill review --optimize ./scientific-skills/scikit-survival/SKILL.mdQuality
Discovery
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.
This is an excellent skill description that clearly defines its scope, provides explicit trigger guidance with a 'Use this skill when' clause, and includes comprehensive domain-specific terminology that practitioners would naturally use. It is highly distinctive and unlikely to conflict with other skills due to its specific focus on survival analysis with scikit-survival.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | Lists multiple specific concrete actions: survival analysis, time-to-event modeling, fitting Cox models, Random Survival Forests, Gradient Boosting models, Survival SVMs, evaluating with concordance index or Brier score, handling competing risks. | 3 / 3 |
Completeness | Clearly answers both what ('comprehensive toolkit for survival analysis and time-to-event modeling in Python using scikit-survival') and when ('Use this skill when working with censored survival data, performing time-to-event analysis, fitting Cox models...') with an explicit 'Use this skill when' clause. | 3 / 3 |
Trigger Term Quality | Excellent coverage of natural terms a user would say: 'survival analysis', 'time-to-event', 'censored survival data', 'Cox models', 'Random Survival Forests', 'concordance index', 'Brier score', 'competing risks', 'scikit-survival'. These are the exact terms practitioners use. | 3 / 3 |
Distinctiveness Conflict Risk | Highly distinctive niche — survival analysis with scikit-survival is a very specific domain. The triggers (Cox models, survival SVMs, concordance index, Brier score, competing risks) are unlikely to conflict with general ML or statistics skills. | 3 / 3 |
Total | 12 / 12 Passed |
Implementation
57%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
The skill is highly actionable with excellent code examples and good progressive disclosure through well-organized reference files. However, it is significantly too verbose — it explains survival analysis concepts Claude already knows, has redundant sections (best practices and pitfalls overlap), and includes a lengthy 'When to Use' section that duplicates the frontmatter description. Workflows lack validation checkpoints for data quality and model convergence.
Suggestions
Remove the 'Overview' paragraph explaining what survival analysis is and the 'When to Use This Skill' section — Claude knows these concepts and the frontmatter description covers this.
Consolidate 'Best Practices' and 'Common Pitfalls' into a single concise section, removing duplicated advice (e.g., standardizing features and Uno's C-index appear in both).
Add validation checkpoints to workflows: verify data quality (check event counts, negative times) before fitting, check model convergence after fitting, and include error handling for common failure modes.
Move the detailed model type listings and decision tree to a reference file, keeping only a brief summary in the main skill to reduce token usage.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The skill is excessively verbose at ~350+ lines. It explains what survival analysis is, when to use it, and includes a 'When to Use This Skill' section that restates the description. The overview explains concepts Claude already knows (what censored data is, what survival analysis aims to do). Best practices and common pitfalls sections have significant overlap. Much content could be trimmed or moved to reference files. | 1 / 3 |
Actionability | The skill provides fully executable, copy-paste ready code examples across multiple workflows. Code includes proper imports, complete pipelines, and specific function calls with real parameters. The model selection decision tree and metric selection guidance (Harrell's vs Uno's based on censoring rate) are concrete and actionable. | 3 / 3 |
Workflow Clarity | Four workflows are clearly sequenced with numbered steps and executable code. However, there are no validation checkpoints or error recovery steps — no checks for data quality issues, no verification that models converged, no feedback loops for when fitting fails or produces poor results. For survival analysis (where data issues like insufficient events per feature can silently produce bad results), validation steps are important. | 2 / 3 |
Progressive Disclosure | The skill has a clear overview structure with well-signaled one-level-deep references to six specific reference files (cox-models.md, ensemble-models.md, etc.). Each reference is clearly described with its scope. The main file provides enough context to get started while pointing to detailed guides for deeper topics. | 3 / 3 |
Total | 9 / 12 Passed |
Validation
90%Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.
Validation — 10 / 11 Passed
Validation for skill structure
| Criteria | Description | Result |
|---|---|---|
metadata_version | 'metadata.version' is missing | Warning |
Total | 10 / 11 Passed | |
- Repository
- K-Dense-AI/claude-scientific-skills
- Commit
25e1c0f
- 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.