Kinetic Multicam — Development skill for Claude Code

Kinetic Multicam

Turn ONE talking-head take into a **kinetic supers + rapid camera reposition** prompt for video-to-video AI generators.

You film a single clip on a single camera. The skill finds the beats of your speech, fills a battle-tested prompt template with 4–6 timeline blocks (scaled to the clip's length) — each one a split-second, visibly motion-blurred camera whip that locks into a stable frame, plus a bold kinetic super (motion-graphics text, not a caption) cut from the exact words you speak in that window. Same face, same room, same voice, same lip sync. No editing timeline, no After Effects.

Validated on **Seedance 2.0 Fast** (2026-07). Google Omni also runs the template, but with clearly weaker results — prefer Seedance 2.0 Fast.

How it works

1. Drop your talking-head video in a folder and ask for the kinetic multicam prompt.
2. The skill runs a preflight check (scripts/check_env.py) and, if anything is missing, explains what it is for and asks permission to install it — so it's plug-and-play.
3. It transcribes your video word-level (Whisper, runs locally) to find exactly where each phrase starts and where you breathe.
4. It suggests 4–6 timeline blocks on those beats (boundaries land on phrase starts, inside breaths, never mid-word) and drafts the supers — verbatim crops of the key words you actually say in each block, in your spoken language.
5. You confirm or adjust the timestamps, the super texts and the camera positions (tight close-up, left orbit, higher crane, wide room reveal, back to the original framing — with presets to swap any of them; every position must contrast with the source framing and its neighbors, and positions that match the phrase's meaning win). You can also opt into a frame-1 hook super over the opening hold.
6. It fills the frozen template, self-checks it (scripts/verify_prompt.py) and saves a ready-to-paste prompt. Upload your clip into Seedance 2.0 Fast, select it as the source footage, paste, gen