What problem does it solve? Turning a written script into a short video normally requires templates, editing tools, or motion-graphics skills. This Skill lets an AI agent convert a paragraph of text into a hand-drawn, marker-on-paper style video by selecting a matching visual card for each sentence and rendering the result to MP4. ## Core Features & Use Cases - Semantic card selection: Picks from 64 hand-drawn scene cards (flows, comparisons, charts, metaphors, UI frames) sentence by sentence, with rules preventing adjacent repeats and clustered high-energy shots. - Deterministic hand-drawn rendering: Builds frames from 65 rough.js stroke primitives with seeded randomness, Excalifont/Xiaolai fonts, and three aspect ratios (9:16, 16:9, 1:1). - Five pre-render quality gates: Lint scripts check portability, browser runtime errors, scene discipline, motion timing, and pixel-level frame audits before any MP4 render is allowed. - Optional voiceover and captions: Supports a vendor-neutral TTS contract with word-level hand-drawn caption highlighting. - Use Case: Give your agent a product announcement paragraph; it produces a storyboard, builds one HTML frame per shot, passes all quality gates, and delivers a vertical MP4 ready for short-video platforms. ## Quick Start Ask your agent to turn a paragraph of text into a hand-drawn style video, for example: make a hand-drawn video explaining what my CLI tool does, its three main features, and how to install it.
