scikit-survival
scikit-survival 0.28.0 enables you to construct, tune, and assess survival workflows for right-censored outcomes using Cox models, ensemble methods, and kernel-based approaches. The skill enforces leakage-safe data splits, preprocessing pipelines, and proper metric selection—discrimination, calibration, and prediction error—while handling competing risks through nonparametric cumulative incidence.
scikit-survival lets you build and evaluate right-censored survival models using Cox regression, forests, boosting, and SVMs with leakage-safe pipelines.
AI-generated summary based on this skill's SKILL.md
Decision gist · record as of 2026-07-28
scikit-survival lets you build and evaluate right-censored survival models using Cox regression, forests, boosting, and SVMs with leakage-safe pipelines. scikit-survival 0.28.0 enables you to construct, tune, and assess survival workflows for right-censored outcomes using Cox models, ensemble methods, and kernel-based approaches. The skill enforces leakage-safe data splits, preprocessing pipelines, and proper metric selection—discrimination, calibration, and prediction error—while handling competing risks through nonparametric cumulative incidence.
Use it when
- scikit-survival enforces leakage-safe data splits and preprocessing pipelines to prevent information from the test set influencing training.
- scikit-survival 0.28.0 computes discrimination metrics like IPCW concordance index and time-dependent AUC, calibration assessments.
Verify before relying
Read SKILL.md below before installing (13 files). Open directory: indexed for reading, not audited.
Install
K-Dense-AI/scientific-agent-skills/scikit-survival · repository language: Python
Open directory. Skills are indexed for reading, not audited. Review a skill's body before installing it.
Frequently asked questions
AI-generated answers based on this skill's SKILL.md and metadata
What survival analysis models does scikit-survival support?
scikit-survival 0.28.0 enables you to build and evaluate right-censored survival models using Cox proportional hazards, random survival forests, gradient boosting survival analysis, and Cox SVM. Each approach handles censoring natively and integrates with scikit-learn pipelines for leakage-safe preprocessing and nested model selection workflows.
How do you prevent data leakage in survival preprocessing?
scikit-survival enforces leakage-safe data splits and preprocessing pipelines to prevent information from the test set influencing training. Use nested cross-validation for hyperparameter tuning and ensure all transformations—imputation, scaling, feature selection—are fit only on training folds before applying to held-out data.
What censoring-aware evaluation metrics does scikit-survival compute?
scikit-survival 0.28.0 computes discrimination metrics like IPCW concordance index and time-dependent AUC, calibration assessments, and prediction error measures such as Brier score. These metrics account for right censoring and competing risks, enabling rigorous model comparison on survival outcomes.
How do you handle competing risks and estimate cumulative incidence?
scikit-survival handles competing risks by computing cumulative incidence functions for each event type nonparametrically. This allows you to estimate the probability of a specific event occurring before competing events, essential for realistic risk prediction in multi-event survival settings.
Can scikit-survival integrate with scikit-learn pipelines for survival prediction?
Yes. scikit-survival models follow scikit-learn's estimator interface and work seamlessly in Pipeline objects. This integration enables you to chain preprocessing, feature engineering, and survival model fitting while maintaining leakage-free cross-validation and hyperparameter tuning.
Does scikit-survival offer command-line tools for survival workflows?
scikit-survival 0.28.0 includes local CLI tools for survival data validation and reporting, supporting automated workflows for preprocessing checks, model evaluation, and result generation outside interactive notebooks.
SKILL.md
Rendered from the published skill. Quoted content, verbatim.
scikit-survival
Scope
Use this skill for scikit-survival 0.28.0 workflows involving:
- right-censored structured outcomes;
- Cox PH, Coxnet, IPC ridge, survival trees, forests, boosting, and SVMs;
- discrimination, prediction error, calibration-oriented checks, and time-dependent prediction;
- nonparametric cumulative incidence with competing risks;
- scikit-learn pipelines, nested model selection, and reproducible reports.
scikit-survival primarily models right-censored outcomes. Its built-in competing-risk support is nonparametric cumulative incidence; it does not provide Fine-Gray regression. Do not present model output as clinical advice, causal evidence,
(truncated - see the full file via the links below)
File tree — 13 files
skills/scikit-survival/SKILL.md
skills/scikit-survival/references/competing-risks.md
skills/scikit-survival/references/cox-models.md
skills/scikit-survival/references/data-handling.md
skills/scikit-survival/references/ensemble-models.md
skills/scikit-survival/references/evaluation-metrics.md
skills/scikit-survival/references/svm-models.md
skills/scikit-survival/scripts/_common.py
skills/scikit-survival/scripts/competing_risk_cif.py
skills/scikit-survival/scripts/evaluate_survival_metrics.py
skills/scikit-survival/scripts/model_report.py
skills/scikit-survival/scripts/train_survival_model.py
skills/scikit-survival/scripts/validate_survival_csv.py
Let your AI agent find skills like this
Example. Real query, live index.
You found this page by searching. An agent finds it by wishing: SkillFed indexes 56,283 agent skills by what they can do, searchable in plain language.
wish › “Build and evaluate right-censored survival models with Cox, forests, or boosting”
Give your agent the search over MCP, or paste the wish link into any chat. No install? Search from any chat →
Related skills
Scikit Survival Analysis extends scikit-learn for right-censored survival data, offering Cox proportional hazards (including elastic net variants), ensemble methods like Random Survival Forests and Gradient Boosting, and survival SVMs. It provides censoring-aware evaluation metrics and non-parametric survival curve estimation.
scikit-survival is a Python library for survival analysis that handles censored data in time-to-event modeling. It offers Cox proportional hazards models, ensemble methods like Random Survival Forests and Gradient Boosting, and Survival SVMs, with evaluation metrics including concordance index and Brier score. Use it for competing risks analysis, Kaplan-Meier estimation, and any survival analysis workflow.
This skill handles survival analysis and time-to-event data common in clinical research, including Kaplan-Meier estimation, Cox proportional hazards regression, and competing risks modeling. It also covers longitudinal data analysis, diagnostic accuracy assessment, and missing data strategies tailored to biomedical studies.
Lifelines is a Python survival analysis library built for medical research and epidemiology. It models time-to-event outcomes like disease progression or recovery, properly accounting for censored data from patients who leave studies early. Use it to estimate survival curves, compare treatment groups, and identify risk factors through Cox proportional hazards regression.
This skill teaches statistical methods for analyzing when events occur, accounting for incomplete observations where some subjects haven't experienced the event yet. You'll learn Kaplan-Meier estimation for survival curves, Cox regression for modeling risk factors, and log-rank tests for comparing groups—with practical Python implementations using lifelines.
Regression Modeling trains predictive models to forecast continuous outcomes and quantify relationships between variables. It supports linear, polynomial, ridge, lasso, and robust regression approaches with comprehensive evaluation metrics and cross-validation.