update: add minimax-chart skill

[mvanhorn]

Summary

Adds a minimax-chart skill that generates publication-ready data visualizations from CSV, JSON, or inline data.

What it does

• Accepts CSV, TSV, or JSON data files
• Auto-detects the best chart type based on data shape (categories + values -> bar, datetime + numeric -> line, etc.)
• 6 chart types: bar, line, scatter, pie, histogram, heatmap
• 4 style presets: modern (clean/minimal), vibrant (bold colors), academic (serif/grayscale-friendly), dark (dashboard-style)
• Outputs PNG, SVG, or PDF at configurable DPI

Why

The skills repo covers document generation (PDF, PPTX, XLSX, DOCX) and development guidance, but nothing for turning raw data into charts. Data visualization is one of the most common developer/analyst tasks. This fills that gap with a matplotlib-based chart generator that auto-detects chart types and applies consistent styling.

Structure

skills/minimax-chart/
  SKILL.md                          # Skill definition with 7-step workflow
  scripts/
    chart_create.py                 # Chart generator with auto-detection + styles
  references/
    requirements.txt                # matplotlib, pandas
    chart-types.md                  # Chart selection guide and style guidelines

The chart_create.py script handles data loading (CSV/JSON/TSV), chart type auto-detection, style application, and rendering. Each style preset defines colors, background, grid, text color, and font family.

Test plan

• Verify SKILL.md frontmatter parses correctly
• Run python3 -m py_compile scripts/chart_create.py (passes)
• Test with sample CSV: python3 scripts/chart_create.py sample.csv -o chart.png

This contribution was developed with AI assistance (Claude Code).