What problem does it solve? Programmatic video tools handle rendering but not direction, so code-generated videos often look like slideshows: reveals miss the voiceover, cursors are invisible, sound effects are inaudible, and every output looks the same. This Skill encodes the motion-design craft — ten laws, a fourteen-step pipeline, fourteen shot constructions, and thirty-four measured traps — so an agent can direct and render a deterministic launch film from code. ## Core Features & Use Cases - Word-locked sync: every reveal is cued to the measured start time of the word it illustrates, using a word-level transcript of the actual voiceover file. - Deterministic rendering pipeline: scripts assemble the film index, inject transitions, wire audio and film grade, master to broadcast loudness (−14 LUFS), and verify cues in the delivered MP4. - Creative direction discipline: forces three competing visual directions before building, with an approved-figures list so no invented metric ever appears on screen. - Use Case: Ask for a 45-second launch video for your product; the Skill writes the brief, script, and voiceover, derives frame timings from the audio, builds one HTML frame per shot with a camera rig and cursor, then renders and masters the final MP4. ## Quick Start Ask the agent to make a 45-second launch video for your product at your website URL, and it will take it from there.
