Claude Code Ships a /dataviz Skill for Charts

Claude Code added a /dataviz skill that checks chart color palettes for contrast and accessibility problems before you write matplotlib, Recharts, d3, or SVG code, per the changelog. Models get these colors wrong by default: a categorical palette that looks fine to the person prompting it can be unreadable to someone with color vision deficiency, and most teams don't catch it until a user complains.

The skill isn't just written guidance. It ships with a runnable palette validator you invoke before generating a chart, and it checks contrast and categorical-color problems in both light and dark themes.

Run it ahead of the chart code, not after. That's the whole pitch: catch the color problem at generation time instead of during a design review, or worse, after launch.

Teams that skip /dataviz and generate charts straight from a prompt will ship at least one contrast bug that a user catches before a reviewer does. Accessible-by-default is cheaper than fixing colors after launch, and now there's a command that makes default mean something instead of a checklist item nobody runs.

It's listed in Claude Code's changelog as of July 2, 2026, while most chart-generation workflows still skip a color-accessibility check entirely.