What problem does it solve? Building visual whiteboard graphics for YouTube videos by hand is slow, and reading content out of an existing .tldr board file requires knowing its internal JSON structure. This Skill provides both an authoring workflow for programmatic shape creation and an inspection workflow for parsing board content. ## Core Features & Use Cases - Programmatic Authoring: Create text, geo, and arrow shapes via the tldraw Editor API in a React app, with sections offset horizontally for panning during recording. - Design Conventions: Enforces David's style rules such as xl uppercase headlines, color-coded concepts, point-based arrows, and zoom-to-fit on load. - Board Inspection: Parse .tldr JSON files, walk the ProseMirror richText tree, sort shapes by position, and isolate a single column of content. - Use Case: Given a video outline with three sections of bullets, generate a new src/videos/<video-name>.tsx file that renders color-coded sections side by side, then review it live at localhost:5173. ## Quick Start Ask the agent to build a tldraw whiteboard graphic from your video outline, or ask it to read the contents of a specific column from an existing .tldr file.