catalog-footage

I'd like you to add a skill to Wideframe called "catalog-footage".

What it should do:
Transcribe and content-analyze every clip, then deliver one coverage-checked breakdown.

Before you run anything, ask me to confirm:
- Which folder or workspace holds the clips, and should subfolders be scanned?
- Transcribe every clip, or only clips with spoken audio?
- Output format and where to save it (table, markdown, spreadsheet)?
- Which fields per clip (transcript, scene type, shot type, summary)?

Then do the work:
1. Enumerate every media file under the target with media-info-batch and dedupe.
2. Submit media-analyze-content-batch for scene/shot/summary on every clip.
3. Submit media-transcript-get-batch for every clip that has audio.
4. Poll with job-poll until no batch is still running.
5. Compile one row per clip with the confirmed fields and flag any unreadable file.

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 output has exactly one row per discovered clip (rows equal files found).
- Every clip has an analysis result, and every clip with audio has a transcript.
- No batch job is still pending at hand-off.
- Unreadable or skipped files are listed explicitly.

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 "catalog-footage" so I can run it again by saying /catalog-footage.