メニュー
Systematic exploratory data analysis. Activate when a dataset needs profiling — structure check, nulls, outliers, distributions, correlations — before deeper analysis begins.
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
SOC 職業分類に基づく
Create standardized metadata for data assets. Use when documenting new datasets, building data catalogs, improving data discoverability, or creating data dictionaries for teams.
Structure analysis approach before starting work. Use when receiving new analysis requests, breaking down complex questions into steps, or planning iterative analysis workflows.
Comprehensive data quality assessment against business rules, schema constraints, and freshness expectations. Activate when validating production pipelines, onboarding new data sources, or when stakeholders report data quality concerns.
Identify distinct customer or user segments based on behavior, attributes, or value. Activate when you need to answer "who are our best customers?" or "what distinct groups exist in our user base?" and need data-driven profiles to inform strategy.
Create effective, publication-ready data visualizations. Use when choosing chart types, designing presentation visuals, building dashboard charts, or applying visual design best practices to data output.
Create a clear, transparent explanation of analytical methodology for any audience level. Activate when you deliver findings that require the audience to trust the method — A/B tests, attribution models, forecasts, statistical analyses, or anything where "how did you get that?" is a likely question.
| name | programmatic-eda |
| description | Systematic exploratory data analysis. Activate when a dataset needs profiling — structure check, nulls, outliers, distributions, correlations — before deeper analysis begins. |
scripts/data_overview.py to get row count, dtypes, memory usage, and a sample. Confirm grain (what one row represents).scripts/null_profiler.py; compare output against thresholds in references/quality_thresholds.md and flag columns above limits.scripts/outlier_detector.py (IQR + z-score) on numeric columns; document flagged values and decide: real signal or data error?scripts/distribution_summary.py for descriptive stats and univariate histograms on each numeric column.scripts/correlation_explorer.py; flag pairs with |r| > 0.8 as potential multicollinearity or redundancy.references/eda_checklist.md and confirm each item before declaring the dataset profiled.assets/eda_report_template.md with full profiling output; distil top issues into assets/findings_summary.md.For pattern recipes (e.g. polars vs pandas equivalents, chunked reads for large files), see references/pandas_polars_recipes.md.
references/quality_thresholds.md)assets/eda_report_template.md (filled) — full profiling report with per-column statsassets/findings_summary.md (filled) — top 3–5 quality issues and recommended next steps