data-feature

Generate and validate predictive features for tabular and time-series datasets.

Updated Mar 3, 2026

Search Tags:#time-series #scaling #shap #feature-engineering #leakage-prevention #feature-selection

Ready to run

npx skills add https://github.com/mutsumi-yamamoto/claude-data-analysis-marketplace --skill data-feature

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: data-feature
Source: https://github.com/mutsumi-yamamoto/claude-data-analysis-marketplace/tree/main/plugins/data-analysis/skills/data-feature
Command: npx skills add https://github.com/mutsumi-yamamoto/claude-data-analysis-marketplace --skill data-feature

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Converts domain hypotheses and raw project data into a validated set of predictive features while preventing common pitfalls such as target leakage and multicollinearity, and producing an auditable feature engineering record.

Core Features & Use Cases

Feature ideation from hypotheses: map analysis_context.md hypotheses to concrete candidate features and data sources.
Transformations & encodings: guidance for numeric transforms, binning, one-hot/target encoding, scaling, and time-series lags/rolling features.
Safety checks & validation: explicit leakage prevention, train/test split handling, VIF checks for multicollinearity, and recommended importance estimation (SHAP/permutation).
Operational output: produces a documented execution log and a saved report for downstream modeling (data/docs/05_feature_engineering.md).

Quick Start

Use the data-feature skill to generate candidate features from analysis_context.md and project data, perform leakage and VIF checks, and save the feature engineering report to data/docs/05_feature_engineering.md.

Frequently Asked Questions about data-feature

High-intent search queries and answers about installing and using this skill.

FAQPage Schema

How do I prevent target leakage when engineering features for time-series data?▼

To prevent target leakage during feature engineering, enforce strict train/test split rules and perform leakage checks before applying any transformations like lags or rolling features. This validation ensures future data does not incorrectly influence predictive modeling.

What is the best way to generate predictive features from raw project data?▼

The best way to generate predictive features is mapping domain hypotheses to concrete candidate features, then applying transformations like encoding, scaling, and time-series lags. This process outputs an auditable feature engineering report for downstream modeling.

How do I check multicollinearity before applying feature transformations?▼

You check multicollinearity by running Variance Inflation Factor (VIF) checks on your candidate features. Performing these checks before applying transformations helps identify and flag redundant predictive features early in the workflow.

Can I estimate feature importance using SHAP for tabular datasets?▼

Yes, you can estimate feature importance for tabular datasets using SHAP or permutation importance methods. This validates the predictive value of generated features after applying necessary encodings and safety checks.

Does feature engineering work without a CRISP-DM workflow context?▼

Feature engineering can technically work without it, but this process specifically reads analysis_context.md as the single source of truth within a CRISP-DM workflow to propose candidate features and document the execution log.