Use when converting any raw text (lyrics, screenplay, brief) into an indexed shot list and chapter summary for a video production.
Converts any raw text input into a structured chapter and indexed shot list. The agent reads the input text, condenses it, divides it into shots, names each shot, and identifies characters and scenes per shot.
chapter.json#rawText--base-dir <path> — project root--chapter-dir <path> — e.g. <base-dir>/chapters/chap_abcNone. This is the first skill in the production pipeline.
condensedText.pending).chapter.json:
# chapter.json is written manually by the agent — no script for this file
shot-list.json to --chapter-dir.source .env && node skills/breaking-down-video-script/scripts/validate-shot-list.js \
--file <chapter-dir>/shot-list.json
<chapter-dir>/chapter.json — chapter metadata with raw and condensed text<chapter-dir>/shot-list.json — indexed shot arraysource .env && node skills/breaking-down-video-script/scripts/validate-shot-list.js --file <path>
Exits 0 on valid, 1 on invalid (error to stderr).
Log to {project_dir}/logs/production.jsonl. See skills/lib/logging-guide.md for schema.
On invocation — key inputs: script_path, chapter_id
On completion — key outputs: shot_list_path, shot_count, chapter_summary_path