multicam-sync

I'd like you to add a skill to Wideframe called "multicam-sync".

What it should do:
Sync camera angles by audio into a multicam sequence, each angle on its own track.

Before you run anything, ask me to confirm:
- Which clips or angles belong to the same take?
- Sync by audio waveform, embedded timecode, or a clap/marker?
- Acceptable sync tolerance (for example one frame)?
- Each camera and mic on its own labeled track?

Then do the work:
1. Group the clips by take.
2. Extract audio peaks for each angle with media-audio-peaks.
3. Cross-correlate with media-sync-corr to get an offset per angle.
4. Create the sequence and place each angle on its own track at its offset.
5. Spot-check alignment at sampled points and report each offset.

Do not grade your own work. When the output is ready, hand it to a separate verification subagent that has its own context window. Give it the output, the choices I confirmed, and the pass criteria below. It reviews the quality, writes specific feedback, and returns PASS or FAIL.

Pass criteria:
- Every angle has a computed numeric offset that was applied.
- Cross-correlation confidence is above threshold for each pair, or the pair is flagged unverified.
- Each camera and audio source sits on its own track.
- Every supplied clip is placed in the sequence.

If it returns FAIL, apply the feedback, redo the affected work, and verify again with a new subagent. Repeat until it returns PASS, then show me the result.

Once it works, save it as a reusable skill named "multicam-sync" so I can run it again by saying /multicam-sync.