Subtitle Packager

Use When

• A stable timed transcript already exists and the job is subtitle chunking, SRT/ASS generation, readable caption packaging, or editor handoff.
• This skill packages transcripts. It must not call speech-to-text or video analysis models.

Do Not Use When

• The task belongs to ideation, QA, or another released skill listed in the handoff section.
• Required inputs are missing and guessing would change the result.

Input Boundary

• Primary input is normalized-transcript.json from audio-transcription or video-transcription with stable timing.
• If timing is missing or invalid, stop and say so plainly. Do not invent timing unless the user explicitly asks for heuristic timing.
• Preserve source transcript path, subtitle mode, output path, and source basis in handoff notes or intermediate JSON.