If you're preparing figures for journal submission, this handles the tedious parts: journal-specific sizing (Nature's 89mm single column vs Science's 55mm), colorblind-safe palettes like Okabe-Ito, proper DPI for different figure types, and export to PDF/EPS/TIFF. It includes matplotlib style presets for major journals and helper scripts that check if your figure meets publication requirements before you submit. The skill emphasizes what actually gets papers rejected: using JPEG instead of TIFF, forgetting that red-green heatmaps fail for colorblind reviewers, or submitting line art at 72 DPI. The multi-panel layout examples with proper panel labeling are solid, and the grayscale testing reminder alone could save you a revision round.
npx skills add https://github.com/davila7/claude-code-templates --skill scientific-visualization