Agent Skills › 0xsline/OpenChatCut › storyboard-shot-breakdown

storyboard-shot-breakdown

GitHub

分析视频镜头，从构图、焦距、运镜、剪辑和叙事功能五个维度解读导演视觉决策，生成分镜参考文档。

src/agent/skills/storyboard-shot-breakdown/SKILL.md 0xsline/OpenChatCut

Trigger Scenarios

用户希望进行逐镜头电影分析需要导演逻辑或摄影拆解请求从视频中生成风格化的分镜参考

Install

npx skills add 0xsline/OpenChatCut --skill storyboard-shot-breakdown -g -y

More Options

Non-standard path

npx skills add https://github.com/0xsline/OpenChatCut/tree/main/src/agent/skills/storyboard-shot-breakdown -g -y

Use without installing

npx skills use 0xsline/OpenChatCut@storyboard-shot-breakdown

指定 Agent (Claude Code)

npx skills add 0xsline/OpenChatCut --skill storyboard-shot-breakdown -a claude-code -g -y

安装 repo 全部 skill

npx skills add 0xsline/OpenChatCut --all -g -y

预览 repo 内 skill

npx skills add 0xsline/OpenChatCut --list

SKILL.md

Frontmatter

{
    "name": "storyboard-shot-breakdown",
    "description": "Break down each shot and turn the analysis into a storyboard reference. Use when the user wants shot-by-shot film analysis, director logic, cinematography breakdown, or a storyboard-style reference from a video."
}

Storyboard Shot Breakdown

Use this skill to analyze a video shot by shot and produce a storyboard-style reference that explains the director's visual decisions.

Core Philosophy

Do not merely describe what is visible. Deduce why each visual choice was made.

Analyze each shot across five director-decision dimensions:

Dimension	Core question
Composition	What is the viewer forced to look at? What pressure does space, symmetry, depth, or framing create?
Focal length	What psychological distance does the lens create? Does it compress, expand, isolate, or invade?
Movement	Why does the camera move or stay still at this exact moment? What changes before and after?
Cut	Why cut here instead of earlier or later? Is the driver emotion, story, rhythm, eye trace, action, or sound?
Narrative function	What new information does this shot deliver: setup, turning point, emphasis, concealment, revelation?

Workflow

Read project state and source media.
Read transcript or audio context when available so narrative interpretation is grounded.
Detect or mark shot boundaries with available OpenChatCut/media tooling.
Extract representative frames for each shot in batch when possible.
Inspect frames before writing analysis. Do not invent details that are not visible.
For high shot counts, tell the user the count and offer output scope choices before spending effort.
Analyze each selected shot using the five dimensions.
Summarize the overall visual rule of the scene in 1-2 concise sentences.
Generate a storyboard reference image or structured analysis artifact if the user wants a visual output.

Internal Reasoning

Use a counterfactual test: if the conventional alternative were used, what emotion or information would be lost? The answer is usually the director's motive.

Use this reasoning to improve the analysis, but do not output the counterfactual test as a separate section unless the user asks.

Output Format

Per shot:

S{N} - {scale} · {lens/mood} · {movement} · {timecode} · {duration} · {narrative function}

Composition: [what attention is forced onto and what pressure the frame creates]
Focal Length: [psychological distance and dramatic effect]
Movement: [why it moves or stays still, and what changes]
Cut: [why the cut lands here, including rhythm/eye trace/sound when relevant]
Narrative: [what story information this shot adds]

Overall:

Director's Logic: [3-5 sentences summarizing the visual strategy]
Visual Rule: [1-2 sentences the user can reuse as a shooting/editing reference]

Rules

Write from visible evidence and available transcript/audio context.
Do not borrow details from adjacent shots unless clearly stated as sequence-level analysis.
Keep shot numbering stable.
User owns scope trade-offs: if output must be reduced, propose choices.
Prefer current OpenChatCut/media tools over hardcoded local commands. Use external scripts only when they are available and clearly help.