AI short-drama production skill collection: character bibles, adaptation outlines, art bibles, screenplays, storyboards. Runs in Claude Code & codex.
shuohao-skills is a collection of agent skills for turning a novel into production materials for an AI short-drama pipeline, including character profiles, outlines, art references, scripts, and storyboards. It is for creators using Claude Code or Codex to plan and prepare AI-generated short videos. The catalogue add-ons implement the separate stages of this workflow and can also assemble their reports.
This repository also configures its own agents. See what shuohao-skills tells them →
Skill Claude Code
A planning kit for creating consistent AI-generated artwork for a short series, covering locations and story-important props. It produces scene anchors, lighting variations, empty-scene prompts, and prop states.
Skill Claude Code
A workflow that reads a novel or short story and creates character profiles, visual prompts, voice prompts, character reference images, and an interactive report. It supports multiple report languages and visual styles.
Skill Claude Code
A Chinese-language tool for turning a novel into a short-drama outline package. It produces adaptation notes, character and appeal-point tables, episode summaries, and an asset list, with scripted checks for quality rules.
Skill Claude Code
A tool for turning an outline for a short AI-produced drama into structured episode scripts. Each script separates actions from individual lines of dialogue and includes timing information.
Skill Claude Code
A storyboard tool for turning a short-drama script into video shots and keyframes. It divides each video segment into cuts and links every cut to specific parts of the script.
Skill Claude Code
A character-reference image workflow that turns a written character description into a consistent set of labelled views, starting from one full-body reference image.
Skill Claude Code
A tool for deciding where characters are positioned at the start of each scene in an AI short video. It uses scene images and script data to record who is present, whether each person sits or stands, and which way they face.
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: