plotting

name	plotting
description	Use this skill whenever creating matplotlib/seaborn figures for academic papers, research reports, or publication-quality plotting tasks using paperops.plotting. Trigger when the user mentions plots, figures, charts, graphs, or visualizations in a research/paper context.

Plotting

Use paperops.plotting to build publication-quality figures for papers and research reports.

Use This Skill When

• the task is to create or revise figures that belong in a paper, report, poster, or appendix
• the user needs chart selection, figure hierarchy, or publication-oriented styling guidance
• a slide needs a real research figure rather than a schematic diagram

Do Not Use This Skill When

• the task is to define a presentation-wide visual language or symbol system
• the task is to build the PPT layout itself
• the task is to review a rendered deck for clipping or layout issues

Workflow

1. Lock the figure claim
   • identify the one conclusion the figure must support
   • choose the chart type that best exposes that conclusion
2. Set the figure system
   • call apply_plot_config(...)
   • choose a consistent size with figure(...)
   • use colors(), markers, line styles, and hatches deliberately
3. Build for layered reading
   • make the main result visible at a glance
   • keep exact values, error bars, and units readable without competing with the main claim
4. Enforce academic constraints
   • axis labels include units
   • bar charts start at zero
   • experimental data includes uncertainty where applicable
   • save vector output for papers and raster output only when the destination requires it
5. Handoff outputs
   • return figure file paths that can be embedded into paperops.slides.Image(path=...)

Handoff

• if the figure is for a deck and the deck layout is the next task, hand off the saved asset to slidecraft
• if the user really needs a diagrammatic slide visual rather than a quantitative chart, hand off to visual-language or slidecraft
• if the user asks for codebase-wide checks after figure-related changes, hand off to verify

References

• prefer the repo README and plotting API docs for exact function signatures
• keep the skill focused on figure decisions, not full API duplication