What problem does it solve? Adding styled captions to existing talking-head footage normally requires manual editing, keyframing, and rotoscoping. This Skill automates the entire pipeline locally: it transcribes speech, segments the subject with a matte, and composites captions into the scene without altering the original footage. ## Core Features & Use Cases - 35-style identity catalog: Pick one visual identity (rail, column-flow, or themed VFX) from CATALOG.md; the engine, compiler, and authoring file are derived automatically. - Rail + embed caption model: A verbatim lower-third rail carries most text while scarce peak words are composited behind the subject via matte occlusion. - Deterministic local pipeline: One prepare script runs matting, Whisper transcription, and safe-zone analysis in parallel, followed by JSON authoring, preview-frame QA, and a gated render to final.mp4. - Use Case: Given a 60-second founder update clip, probe the footage, pick the keynote identity, author a small cinematic.json, preview composite frames, and render a captioned video with the climax word embedded behind the speaker. ## Quick Start Ask the AI to add captions to your talking-head video file using the embedded-captions skill and let it recommend an identity from the catalog.
