Seedance Storyboard skill

A dialogue that turns an idea into a storyboard and a prompt for the Seedance 2.0 video model in Jimeng, with six templates and ten examples

  • 312 ★ on the repository
  • Last commit
  • MIT
Skill

Minimal access

Why this level

  • The skill is made of text instructions and templates, with no scripts in the repository
  • The finished prompt goes into Jimeng by hand, so the skill needs no access to your service account
All reasons and checks

elementsix/elementsix-skills

Install

In your terminal, with SkillFoxx CLI

npx skillfoxx add skills/seedance-storyboard

Detects the agents on your machine, shows the access level and pins the version.

Other ways to install

Access and checks

How we review

Minimal accessThe tool runs nothing and makes no network calls.

  • The skill is made of text instructions and templates, with no scripts in the repository
  • The finished prompt goes into Jimeng by hand, so the skill needs no access to your service account

This is third-party code. Review the repository files before installing.

What it does

The seedance-storyboard skill walks from an idea to a finished prompt for Seedance 2.0, the video model in Jimeng (CapCut). In Claude Code it runs as /seedance-storyboard. The agent asks what the clip is about, how long it should be and whether reference material exists, then clarifies style, camera, motion rhythm and sound. The story is then split into second-by-second segments and the skill assembles a prompt in three parts: an overall description with style and frame format, a storyboard with camera, action and light, and a line about sound. Assets are attached with references such as @image1, @video1, @audio1. At the end the skill gives improvement tips and says which references are worth adding. For harder cases it carries recipes: extending a finished clip, replacing a character, inserting a scene between two videos, copying a camera move from someone else's clip. The package includes six storyboard templates, ten complete examples and a short quick-reference card.

Who it is for. Short-video makers who build clips in Jimeng and still write Seedance 2.0 prompts by feel.

Good fit when

  • You have an image in mind, such as a girl dancing at sunset, but no prompt structure
  • A video is built from references: the first frame from a picture, camera motion from another clip, music from an audio file
  • A finished clip must be extended or its hero replaced without rebuilding everything
  • A 15-second ad should be laid out on a ready template and checked against examples

Not a fit when

  • Clips are made in Runway, Kling or another service: the skill writes @ references and limits for Seedance 2.0
  • The references need realistic human faces: Jimeng does not currently accept such material
  • You need the finished video itself: the skill writes the prompt text and generation happens in Jimeng separately

Example request

/seedance-storyboard A girl dances on the beach at sunset, 15 seconds, 16:9, warm light, calm piano in the background

Limitations

  • The skill outputs prompt text; the video is made in Jimeng (jimeng.jianying.com), which needs a service account.
  • Jimeng currently does not support material with realistic human faces, as the README notes.
  • After installing, Claude Code must be closed and reopened, otherwise the /seedance-storyboard command does not appear.
  • The last commit was in March 2026, so Seedance 2.0 rules may have changed. SKILL.md is written in Chinese, the README in English.

How to disable. The seedance-storyboard@elementsix-skills plugin is disabled and removed in the /plugin menu of Claude Code. The elementsix-skills marketplace is listed on the Marketplaces tab of the same menu.

FAQ

How do I install Seedance Storyboard in Claude Code?

Run /plugin marketplace add elementsix/elementsix-skills and /plugin install seedance-storyboard@elementsix-skills, then restart Claude Code and call /seedance-storyboard. Update the plugin through /plugin: the Marketplaces tab, Update marketplace.

How many assets can I attach to a prompt?

Up to nine images, up to three videos with a total length of 2–15 seconds and up to three audio files up to 15 seconds, twelve files at most. Video references use more of the generation quota, so attach first what most affects the frame and the rhythm.

What duration does the skill write a storyboard for?

It asks for a duration from 4 to 15 seconds and defaults to 15. The timeline is split into segments such as 0–3 s, 3–7 s and so on.

Why did the /seedance-storyboard command not appear after installing?

Claude Code must be fully closed and reopened: before the restart the command is not in the list. If it is still missing, check in the /plugin menu that the plugin is enabled.

How do I attach my own pictures and videos?

References such as @image1 assign a role to an asset: first frame, character reference, camera-move sample, music. Jimeng takes two entries: a first frame with text, or a full set of images, video, audio and text.

Discussion

No questions or reviews yet.

Skills

A skill that teaches an agent to write prompts for ByteDance's Jimeng Seedance 2.0 video model: @ references, camera language, timing, templates

SkillMinimal access4.4KRepository stars

The skill breaks a sales video into a storyboard and writes a prompt pack for the Seedance video model in Jimeng.

SkillExtended access120Repository stars

A skill that improves Seedance prompts, plans storyboards and generates video through Atlas Cloud, plus a library of ready prompts

SkillExtended access260Repository stars

A Claude Code skill that turns one sentence into a vertical AI micro-drama, from script and storyboard to video

SkillExtended access367Repository stars

A skill that turns an idea, theme or novel excerpt into a 30-second multi-shot prompt for the Seedance 2.5 video model.

SkillMinimal access139Repository stars

Awesome Seedance 2 Prompts

awesome-seedance-2-prompts

Over 2000 prompts for the Seedance 2.0 video model from ByteDance: cinema, anime, UGC, ads and memes, plus notes on the API and character consistency

PromptMinimal access2.2KRepository stars

Comparison with similar tools

Cards that share the main search query, and how they differ.

ToolOfficialRepository starsMain difference
Seedance Storyboard (this card)no312
Seedance 2.0 Prompt Writing Skillsno4,413a reference on Seedance 2.0 syntax and camera language; this skill guides step by step and supplies storyboard templates
Seedance 2.5 Skillno260improves prompts and generates video through Atlas Cloud; here only the text is prepared
Seedance2 Storyboard Generatorno2,785Seedance2 Storyboard Generator keeps style and characters consistent

README in short

The README is in English. It introduces the skill briefly and explains marketplace installation with a restart warning. Then come the five dialogue steps, two examples, Seedance 2.0 capabilities and a link to Jimeng. The repository layout is shown as a file tree.

README badge

Original SKILL.md

---
name: seedance-storyboard
description: 将任何想法转换成即梦 Seedance 2.0 专业分镜提示词。当用户想要生成视频、制作短视频、创作分镜、使用 Seedance/即梦/剪映 AI 视频时调用。
disable-model-invocation: false
---

# 🎬 Seedance 2.0 分镜提示词生成器

帮助用户将想法转化为专业的 Seedance 2.0 视频分镜提示词。

## Seedance 2.0 核心能力

### 技术参数

| 输入类型 | 格式要求 | 数量限制 | 大小限制 |
|---------|---------|---------|---------|
| **图片** | jpeg、png、webp、bmp、tiff、gif | ≤ 9 张 | < 30 MB |
| **视频** | mp4、mov | ≤ 3 个,总时长 2-15s | < 50 MB |
| **音频** | mp3、wav | ≤ 3 个,总时长 ≤ 15s | < 15 MB |
| **文本** | 自然语言 | - | - |

**混合输入总上限:12 个文件**

> 💡 建议:优先上传对画面或节奏影响最大的素材。有参考视频会消耗更多生成额度。

**核心功能**:

1. **参考图像** - 精准还原画面构图、角色细节、服装样式
2. **参考视频** - 支持镜头语言、复杂动作节奏、创意特效复刻
3. **视频延长** - 平滑延长与衔接,可"接着拍"
4. **视频编辑** - 角色更替、剧情颠覆、片段调整、场景融合
5. **多模态组合** - 图+视频+音频+文本自由组合创作

**基础能力增强**:
- ✅ 物理规律更合理、动作表现更自然流畅
- ✅ 指令理解更精准、风格保持更稳定
- ✅ 支持复杂动作、连续运动等高难度生成任务

**交互方式**:使用"@素材名"指定用途,如"@图片1 作为首帧,参考@视频1 的运镜"

**入口说明**:
- **首尾帧入口**:只上传首帧图 + prompt
- **全能参考入口**:多模态(图、视频、音频、文本)组合输入

---

## 分步引导流程

### 第一步:理解用户想法

Report a problem

Foxx AISeedance Storyboard
Foxx AI

I am Foxx AI and I have already vetted this tool. Ask about install, setup or anything else, and I will keep it simple.