Script to Storyboard

将电影/视频剧本转换为分镜脚本表格。适用于：(1) 将剧本文本解析为结构化分镜，(2) 根据场景描述生成拍摄计划，(3) 提取视觉元素并输出标准分镜表格（镜号、景别/拍摄角度、画面内容、出场人物、场景、音效）。当用户提到分镜、storyboard、镜头脚本、剧本转换时触发。

Security Scan

The skill's files and runtime instructions match its stated purpose (converting a script into a storyboard CSV); it reads a local text file, parses text heuristically, and writes a CSV — no network, secrets, or unusual privileges requested.

Purpose & Capability

Name/description (script → storyboard) match the included code and SKILL.md. The provided Python script implements text parsing, shot detection, and CSV output consistent with the stated features; no unrelated services or credentials are requested.

Instruction Scope

SKILL.md instructs running scripts/convert.py on a local input file and describes manual conversion steps. The instructions only reference local input/output and the provided script; they do not direct the agent to read unrelated system files, call external endpoints, or exfiltrate data.

Install Mechanism

No install spec is present (instruction-only with a bundled script). The code is plain Python with no downloads, package installs, or archive extraction. Risk from installation is minimal.

Credentials

The skill requests no environment variables, credentials, or config paths. The script only reads the input file specified by the user and writes a CSV; no secret access is required.

Persistence & Privilege

Skill is not always-enabled and has default invocation behavior. It does not attempt to modify other skills or system configurations and does not request persistent privileges.

Assessment

This skill appears coherent and low-risk: it reads a user-provided text file, parses it heuristically, and writes a CSV locally. Before using it, avoid supplying sensitive or confidential scripts (the tool will read whatever file you point it at). Be aware parsing is rule-based and may mis-identify names or dialogue (e.g., Chinese-character speaker names aren't fully covered by the simple regex). Also review sample reference content for cultural sensitivity before using or distributing outputs.

将剧本转换为标准分镜脚本表格。

输入格式

剧本应包含以下元素：

• 场景描述（场景编号、时间、地点）
• 人物动作
• 对白
• 镜头指示（如"特写"、"推镜头"等）

输出格式

生成的表格包含以下字段：

• 镜号: 镜头序号 (如 01, 02, 03...)
• 景别/拍摄角度: 如远景、全景、中景、近景、特写，或俯拍、仰拍、平拍
• 画面内容: 画面描述，包含主要视觉元素
• 出场人物: 该镜头出现的角色
• 场景: 拍摄地点和环境
• 音效: 背景音乐、环境音、对白以外的声音

转换脚本

使用 scripts/convert.py 辅助转换：

python3 scripts/convert.py <input.txt> [output.csv]

脚本会自动：

1. 解析剧本结构，识别场景和镜头
2. 提取视觉元素（人物、动作、环境）
3. 生成符合电影制作规范的分镜表格

手动转换流程

当需要手动转换时：

1. 识别镜头边界：查找剧本中的镜头指示词（远景、特写、切入、切出等）
2. 提取场景信息：从场景描述中获取地点和时间
3. 列出出场人物：从动作/对白中识别角色
4. 描述画面内容：用简洁的视觉语言描述画面
5. 标注音效：记录背景音乐、环境音、特殊音效

示例

参考"热带海岸沙滩派对夜景"剧本结构：

• 风格定位：夜景、热闹、氛围感
• 运镜方式：推、拉、摇、移
• 核心画面：全景展示 → 人物特写 → 细节强化
• 镜头分镜逻辑：开场全景 → 中景互动 → 特写情感 → 群体高潮 → 收尾

分镜输出示例：

镜号	景别/拍摄角度	画面内容	出场人物	场景	音效
01	远景/俯拍	全景俯瞰沙滩派对，灯光璀璨	全体	热带海岸沙滩	背景音乐
02	中景/平拍	人群舞动，香槟塔	派对人群	沙滩派对	音乐+欢笑声