---

name: objects-interface description: "Use seaborn.objects for declarative Plot composition with marks, stats, moves, scales, layering, faceting, and output handling." disable-model-invocation: true metadata: disco-role: operating license: BSD 3-Clause

Objects Interface

Use this sub-skill when a task mentions seaborn.objects, so.Plot, marks, stats, moves, scales, declarative layering, or grammar-like plot construction.

Route Here For

• Building a so.Plot(data, x=..., y=..., color=...) and adding layers with .add(mark, stat, move, ...).
• Choosing mark classes such as Dot, Line, Bar, Area, Band, Text, Range, Dash, Path, or their plural variants.
• Choosing stat classes such as Agg, Est, Count, Hist, KDE, Perc, and PolyFit.
• Using move classes Dodge, Jitter, Norm, Shift, and Stack to adjust layer positions.
• Applying Continuous, Nominal, Temporal, or Boolean scales and setting labels, limits, layout, pair/facet specs, and themes.
• Rendering with .plot(), .show(), or .save(...), or directing output to a matplotlib axes/figure with .on(...).

Use Another Sub-skill For

• Classic sns.*plot functions: ../function-interface/SKILL.md.
• FacetGrid, PairGrid, JointGrid, pairplot, or jointplot: ../figure-grids/SKILL.md.
• Palette/theme constructors used outside Plot.theme: ../themes-palettes/SKILL.md.
• Long/wide-form data validation: ../data-utilities/SKILL.md.

Start With

1. Decide whether the request benefits from declarative layering; if a single high-level function is enough, route to function-interface.
2. Build so.Plot with shared data and variable mappings.
3. Add one layer at a time with a mark and optional stat/move objects.
4. Use .facet() or .pair() for small multiples and .scale() for property/value mapping rules.
5. Render explicitly with .plot(), .save(path), or .show().

References

• Object taxonomy and key method signatures: references/api-reference.md.
• Layering/faceting recipes: references/workflows.md.
• Failure recovery: references/troubleshooting.md.
• Shared package API map: ../../references/api-summary.md.

Quick Smoke Check

python sub-skills/objects-interface/scripts/objects_smoke.py --output objects_smoke.png

The helper uses synthetic data, the Agg backend, and no network access.