train-test-splitter

Quick Review

The skill has a clear structure but lacks specific technical details. The description is too generic and doesn't explain what 'train test splitter' actually does (e.g., splitting datasets into training/testing sets, stratification options, random seeds). There's no concrete guidance on parameters (test_size, shuffle, stratify), no code examples or implementation steps, and no validation criteria. A CLI agent would struggle to invoke this effectively without understanding whether it should use sklearn.model_selection.train_test_split, implement custom splitting logic, handle time-series splits, or perform stratified sampling. The novelty is limited since train-test splitting is a straightforward task that typical ML libraries handle easily. To improve: add specific parameters, splitting strategies, code templates, and edge case handling (imbalanced data, time-series, cross-validation).