extract-stills

I'd like you to add a skill to Wideframe called "extract-stills".

What it should do:
Pull still frames from video as images, by interval, scene, or timecode, at a target resolution.

Before you run anything, ask me to confirm:
- Every frame, a fixed interval, one per scene, or specific timecodes?
- Output resolution and image format (PNG or JPG)?
- In-focus frames only?
- Output folder?

Then do the work:
1. Probe fps, resolution, and duration with media-info-batch.
2. Determine the frame set per the rule (use media-analyze-content-batch for scene detection if needed).
3. Extract the frames at the target resolution and format.
4. If in-focus only, cull blurry or duplicate frames.
5. Write the images to the output folder and report the count.

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:
- The frame count matches the chosen rule (for example one per detected scene).
- Every image is at the requested resolution and format.
- If in-focus was requested, the delivered frames are sharp.
- All files exist at the stated output path.

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 "extract-stills" so I can run it again by saying /extract-stills.