K-Dense Data/Viz Selected

This wrapper packages selected Data Analysis & Visualization skills from K-Dense-AI/scientific-agent-skills for Auto-sci-research.

Use it when a task needs:

• exploratory data analysis;
• statistical test selection and reporting;
• assumption diagnostics, effect sizes, uncertainty, or power analysis;
• Matplotlib, Seaborn, NetworkX, Polars, or Dask technical guidance;
• publication-grade scientific figures.

Included Upstream Subskills

Located in subskills/k-dense/:

• exploratory-data-analysis
• statistical-analysis
• matplotlib
• seaborn
• scientific-visualization
• networkx
• polars
• dask

Local Adaptation

Use these upstream skills with Auto-sci-research rules:

1. Start every figure from the claim it must support.
2. Audit units, missingness, outliers, groups, and spatial/temporal coverage before statistical analysis.
3. Use effect sizes and uncertainty, not only p-values.
4. For bibliometric visuals, connect clusters and networks to field evolution, evidence gaps, and policy relevance.
5. Export figures at journal-ready dimensions with colorblind-safe palettes and readable captions.

For domain-specific guidance, also read:

• ../agent-auto-sci-data-viz/references/k_dense_data_viz_mapping.md
• ../agent-auto-sci-data-viz/references/review_bibliometric_figure_system.md

Must Not Do

• Do not make a figure that does not answer a manuscript claim.
• Do not let visual attractiveness replace evidence.
• Do not hide small sample size, missingness, or uncertainty.
• Do not imply causality from descriptive charts.