Token导航 LogoToken导航TokenDH.com
效率只读clawhub未标认证来源可访问clear审计通过

seedance2-storyboard-generatorSeedance2 故事板生成器

Agent Skill

seedance2-storyboard-generator 用于辅助视频、动画、脚本化剪辑和多媒体生成流程,适合在 OpenClaw 中需要整理视频素材、生成脚本或维护合成项目时使用。可结合来源仓库、安装命令和原始 README 继续核验具体用法。安装前建议确认权限范围、维护状态,以及是否会触发联网、命令执行或文件读写。

总安装

27,537

周安装

1,136

GitHub Stars

公开资料未说明

下载量

8,997
OpenClaw

安装说明

本站只整理中文说明和来源信息,不托管安装包,也不代用户安装。

GitHub

来源数

2

许可证

MIT-0

最后核验

2026-05-01

来源状态

来源可访问

安装方式

通过对话安装

复制提示词发给支持本地命令或 Skills 的 AI 助手,先确认命令和权限,再让它执行。

请帮我安装这个 Agent Skill:seedance2-storyboard-generator(Seedance2 故事板生成器)
来源仓库:https://github.com/liangdabiao/seedance2-storyboard-generator
安装命令:
openclaw skills install seedance2-storyboard-generator
安装前请先检查当前环境是否支持对应 CLI,并向我确认将要执行的命令、安装目录、联网范围和文件读写权限;确认后再执行。

命令行安装

复制命令到本机终端执行。该命令会通过 OpenClaw 从第三方来源获取 Skill;本站只展示命令,不托管安装包,也不自动执行。

ClawHubOpenClaw
openclaw skills install seedance2-storyboard-generator

简介

从故事或文章自动生成四幕 AI 视频脚本与分镜提示。

  • 输出包含剧集细分、资产说明与 Seedance 2.0 专用提示。
  • 加速视频策划阶段的结构化内容准备。seedance2-storyboard-generator 属于效率类 Skill,可作为该场景下的辅助能力补充。
  • 通过 clawhub 安装并运行于 OpenClaw 环境。
  • 生成的脚本需人工校验是否符合实际拍摄与合成条件。

SKILL.md

name
seedance-storyboard-generator
description
Professional AI video script and storyboard generator for Seedance 2.0 platform. Use when user asks to: (1) Convert articles/stories into video scripts, (2) Generate Seedance 2.0 storyboard prompts, (3) Plan multi-episode AI video series, (4) Create character/scene/prop generation prompts for image models like Nana Banana Pro. Input can be full novels, articles, or brief story outlines. Output includes four-act script structure, episode breakdown, asset generation prompts, and Seedance 2.0 formatted storyboard scripts.

Seedance Storyboard Generator

Expert AI script and storyboard generation system for creating professional AI video series on Seedance 2.0 platform.

Workflow

Follow this sequential process to convert source material into production-ready video scripts:

1. Analyze Input

Determine input type:

  • Full text: Complete novel/article requiring adaptation and episode segmentation
  • Outline: Brief story concept requiring full script development

Extract core elements:

  • Protagonist(s) and key characters
  • Central conflict and narrative arc
  • Setting/world-building elements
  • Key plot points and emotional beats

Ask clarifying questions if input is ambiguous or incomplete.

2. Confirm Production Parameters

Essential questions to ask:

  1. Visual Style: What visual style? (写实/动画/水墨/科幻/复古/电影感/其他)
  2. Duration: Total runtime? (Standard: 20 episodes × 15s each ≈ 5 minutes)
  3. Target Platform: Aspect ratio? (16:9横屏 / 9:16竖屏 / 2.35:1电影宽屏)
  4. Tone: Overall emotional tone? (史诗/温馨/悬疑/欢快/忧伤等)

Document these parameters for consistent application throughout.

3. Generate Four-Act Script Structure

Structure:

  • Act 1 (起): Episodes 1-5 - Introduction and inciting incident
  • Act 2 (承): Episodes 6-10 - Rising action and complications
  • Act 3 (转): Episodes 11-15 - Climax and confrontation
  • Act 4 (合): Episodes 16-20 - Resolution and conclusion

For each episode include:

  • Episode number and title
  • Duration (standard: 15 seconds)
  • Emotional tone/mood
  • Key plot points
  • Beginning/ending frame description (for continuity)

Output format: Markdown document with clear section headers.

4. Create Asset Generation Plan

Categorize and number all visual assets:

CategoryPrefixExampleDescription
CharactersC01-C99C01 叶青云·正面全身Multiple angles per character
ScenesS01-S99S01 青锋山废墟Key locations
PropsP01-P99P01 青锋剑Important objects

Asset generation prompt format:

### [编号] — [名称]

[Style prefix], [detailed visual description in English], [technical specs]

**Style Prefix Examples:**
- Chinese ink wash painting style mixed with anime cel-shading
- Cinematic photorealistic style with dramatic lighting
- 3D Pixar-style animation rendering
- Sci-fi cyberpunk aesthetic with neon lighting

**Character differentiation:** Use distinct color schemes and visual markers for each character to ensure recognition in the chosen art style.

Output format: Organized list with unique IDs, suitable for copy-pasting into image generators.

5. Generate Seedance 2.0 Storyboard Scripts

For each episode, produce:

a. Asset Upload List

图片1: C01 (角色参考)
图片2: S03 (场景参考)
图片3: P01 (道具参考)

b. Seedance Prompt (Time-axis format)

[风格描述],[画幅比例],[整体氛围]

0-3s画面:[镜头运动],[场景建立],[主体引入]
3-6s画面:[镜头运动],[情节发展],[动作描述]
6-9s画面:[镜头运动],[高潮/冲突],[情绪爆发]
9-12s画面:[镜头运动],[转折/过渡]
12-15s画面:[镜头运动],[结尾/落版]

【声音】[配乐风格] + [音效] + [对白/旁白]

【参考】@图片1 [用途],@图片2 [用途]...

c. Ending Frame Description

  • Document the final frame content for next episode continuity

Camera movement keywords: 推镜头/拉镜头/摇镜头/移镜头/跟镜头/环绕镜头/升降镜头/希区柯克变焦/一镜到底/手持晃动

For episode chaining (Ep 2+): Start prompt with 将@视频1延长15s and upload previous episode as video reference.

Output Files

Generate these deliverable files:

  1. [Title]_剧本.md - Complete four-act script with episode breakdown
  2. [Title]_素材清单.md - All character/scene/prop generation prompts
  3. [Title]_E[XX]_分镜.md - Individual episode storyboard scripts (or combined)

Quality Assurance

Before finalizing:

  • Verify all asset references (@图片X) have corresponding IDs in asset list
  • Check episode-to episode continuity (ending frame → opening frame)
  • Ensure time-axis coverage spans complete 15 seconds
  • Validate that camera movements are feasible and logically sequenced

Reference Material

For detailed Seedance 2.0 prompt patterns, templates, and best practices, see references/seedance-manual.md.

Key reference sections:

  • Templates 1-16 for different video types (叙事/产品/角色/风景/战争/等)
  • Camera movement quick reference
  • Atmosphere keyword library
  • Multimodal reference syntax (@图片X, @视频X, @音频X)

Common Pitfalls to Avoid

  1. Sensitive words: Seedance may reject content with certain terms. Avoid common triggers or use alternative phrasing.
  2. Over-complex prompts: Long prompts (300+ words) may have inconsistent instruction following. Prefer clarity over verbosity.
  3. Missing continuity: Always document ending frames and verify next episode starts with matching scene.
  4. Inconsistent style: Apply same visual style prefix to all asset generation prompts.

适合场景

01

OpenClaw 用户查找和安装 Skill 时

02

用户想查找某类 Agent Skill 时

03

需要根据任务场景推荐可安装能力包时

04

需要对比不同来源的安装命令和来源信息时

能力概览

能力 1

按任务关键词查找相关 Skills

能力 2

展示可复制的安装命令

能力 3

保留来源站点、仓库和原始说明,方便继续核验

能力 4

补充不同宿主或平台的使用分布数据

能力 5

展示第三方安全扫描或审计结果

安装后应在对应宿主中按原始 README 的触发条件使用;具体调用方式请以来源页面和 README 为准。

平台分布

OpenClaw

75.53%
按下载量换算6,795

安全审计

VirusTotal

通过

ClawScan

通过

Static analysis

未展示

权限和风险

只读

该 Skill 主要提供规则、说明或参考内容,本身偏只读;真正读写文件、联网或执行命令仍取决于宿主 Agent 的任务。

安装前确认

本站仅展示第三方公开信息,不托管安装包,不提供自动安装或运行环境。安装前应自行审查源码、依赖和命令行为。当前只有一个来源,正式发布前建议补源仓库或其他目录站核验。

来源信息

继续浏览同类 Skills