视频提示词生成系统V1，可将视频内容拆解为分镜表、角色设定、画面提示词，支持二次创作生成新分镜

视频提示词生成系统V1，可将视频内容拆解为分镜表、角色设定、画面提示词，支持二次创作生成新分镜

Security Scan

This skill is an instruction-only video-to-prompt/shotlist generator and its declared requirements and instructions are consistent with that purpose.

Purpose & Capability

Name/description (video prompt /分镜生成) match the SKILL.md content: analysis mode (from transcript) and creative mode (from keywords). No unrelated resources, binaries, or credentials are requested.

Instruction Scope

The runtime instructions describe producing role sheets, shot lists and AI-drawing prompts from provided text input. They do not instruct reading system files, environment variables, or calling external endpoints beyond typical generation tasks.

Install Mechanism

No install spec and no code files — instruction-only. Nothing will be downloaded or written to disk by the skill itself.

Credentials

The skill declares no required environment variables, credentials, or config paths. The requested capabilities are proportional to the stated functionality.

Persistence & Privilege

always is false and default autonomous invocation is unchanged. The skill does not request persistent system presence or modify other skills; these privileges are appropriate for an instruction-only generator.

Assessment

This skill is internally consistent and requires no credentials or installs. Before using it: (1) avoid feeding sensitive or private video transcripts because the skill will process and output their content; (2) test with non-sensitive examples to verify output format meets your AI-drawing pipeline; (3) note the author/source is unknown—if you require provenance or support, verify the provider separately; (4) because it’s instruction-only there’s no code running on your machine from the skill itself, but be cautious if your agent is configured to forward data to external connectors or services.

视频提示词生成系统 V1

核心能力

1. 视频内容分析：自动拆解角色设定、分镜结构、画面提示词
2. 分镜模板生成：输出标准化分镜表，支持AI绘图
3. 二次创作：基于关键词生成全新分镜方案

使用方式

模式1：视频分析（输入视频转录文本）

你是专业视频分镜分析师，将视频内容拆解为：

• 角色设定表（含外貌、服装、一致性关键词）
• 完整分镜表（镜头序号、画面内容、时长、景别、运镜）
• 每个镜头的AI绘图提示词
• 视频整体结构分析

模式2：新分镜创作（输入主题/关键词）

基于元提示词模板，自动生成：

• 全新角色设定
• 完整分镜脚本
• 可直接用于AI绘图的画面提示词

画面提示词规范

• 使用{角色A}、{角色B}作为角色占位符
• 提示词包含：景别、光线、色彩、构图、动作、细节
• 保证跨镜头角色视觉一致性
• 适配主流AI绘图模型（Midjourney、Stable Diffusion、豆包、千问等）

输出要求

1. 表格化呈现，清晰易读
2. 中文输出，专业简洁
3. 无冗余内容，重点突出
4. 提示词可直接复制使用