FigDown/figdown: Figures as text — one source, two readers. A Markdown-native standard for diagrams that AI agents read as text and humans see as deterministically rendered SVG.

Source-linked topic cluster with 1 signals across related articles, projects, models, papers, and source updates.