What problem does it solve? Creating a video that stays synchronized to a music track requires manual beat detection, timeline editing, and frame-by-frame alignment. This Skill automates that pipeline: it analyzes the audio once, cuts the track into frames at real musical changes, and builds each frame as an HTML composition rendered to an MP4. ## Core Features & Use Cases - Deterministic beat analysis: A Python script (librosa-based) produces a canonical audiomap.json with beat grid, onsets, drum classification, rolls, energy phases, and key moments. - Storyboard-driven workflow: A six-step gated pipeline (setup, analyze, skeleton, plan, build, assemble, render) with user approval checkpoints and validation scripts. - Template and motion-primitive catalogs: Per-frame groups are realized from reusable HTML templates or free-composed motion primitives, with optional user images/videos cut onto the same beat grid. - Use Case: Turn a 30-second high-energy track into a kinetic lyric promo: the analyzer finds the drops and rolls, the planner assigns templates per section, and the renderer outputs a 1920x1080 MP4 with every cut on the beat. ## Quick Start Ask the agent to turn your music file into a beat-synced video, for example: create a lyric video from assets/bgm.mp4 using the music-to-video skill.
