Data Visualization

Master skill for static plots, interactive charts, dashboards, infographics, and generative art. Routes to specialized references by task type.

Task Router

Decision Tree

Output type?
  ART (generative, algorithmic) --> p5.js (references/generative-art.md)
  INFOGRAPHIC (visual storytelling) --> Gemini MCP (references/infographics.md)
  ANIMATED GIF --> PIL + easing (references/animated-visualizations.md)
  EXCEL FILE --> openpyxl (references/excel-charts.md)
  COLOR THEME --> references/color-themes.md
  DATA CHART -->
    Interactivity needed?
      YES + dashboard --> Plotly Dash
      YES + standalone --> Plotly Express
      NO + statistical --> Seaborn
      NO + general --> Matplotlib OO API
    3D? --> Plotly (interactive) or mplot3d (static)
    Maps? --> references/geographic-maps.md
    Networks? --> references/network-graphs.md
    Live data? --> references/realtime-dashboards.md
    Accessibility? --> references/accessibility-compliance.md

Quick Reference

Matplotlib: Always OO interface: fig, ax = plt.subplots(figsize=(10,6), constrained_layout=True). Save: plt.savefig('fig.png', dpi=300, bbox_inches='tight'). Close: plt.close(fig).

Plotly: Two APIs — px (high-level) and go (fine control). Export: write_html() (interactive) or write_image() (static, needs kaleido).

Colormaps: viridis (default), cividis (colorblind-safe), coolwarm (diverging). Avoid jet.

Accessibility: Colorblind-safe palettes always. Never encode meaning with color alone. Add patterns/markers alongside color. Min contrast 4.5:1 text, 3:1 graphical. Alt text for every saved image.

Performance: >100k points → rasterized=True (mpl) or scattergl (Plotly). Always plt.close(fig) after saving.

Layer 3 Skills