
from milimovideo78
Parse screenplays into structured Scene and Shot records, auto-match elements (characters/locations), and generate concept thumbnails for video storyboards.
Transforms plain-text screenplays into generation-ready data structures (Scene and Shot records), enriches them with deterministic element-matching, and coordinates thumbnail generation for storyboard pipelines. The skill explains two parsing pipelines (fast regex and an AI parser), how matches to characters/locations/items are scored and linked, and how thumbnail jobs are queued and emitted to the front end.
Use this skill when you need to: convert screenplay or script text into a structured sequence of scenes and shots for downstream video generation; debug parsing failures between regex and AI parsers; implement deterministic element linking to improve visual conditioning; or integrate thumbnail/job queue logic into a storyboard UI.
Best used by agents with code execution and HTTP tooling (Copilot/Codex-style or agents that can read and modify backend Python services), and by agents familiar with video toolchains and FFmpeg integration.
This skill has not been reviewed by our automated audit pipeline yet.