Seedance 2.0 Prompt Writing Skills is a set of agent skills for creating prompts for ByteDance’s Seedance 2.0 multimodal video-generation model. It is for users of Claude Code, Cursor, Cline, and similar coding agents who need prompts for videos such as advertisements, dramas, music videos, or educational content. The catalogue entries provide this prompt-writing workflow.
Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add dexhunter/seedance2-skill --skill zhgit clone --depth 1 https://github.com/dexhunter/seedance2-skillWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/dexhunter/seedance2-skill/zh)<a href="https://agentmods.dev/skills/dexhunter/seedance2-skill/zh"><img src="https://agentmods.dev/badge/skills/dexhunter/seedance2-skill/zh/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/dexhunter/seedance2-skill/zh"><img src="https://agentmods.dev/badge/skills/dexhunter/seedance2-skill/zh.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00100 | $0.05615 |
| Opus 5.5 | $0.00040 | $0.02246 |
| Sonnet 5.5 | $0.00020 | $0.01123 |
| Haiku 4.5 | $0.00010 | $0.00562 |
Grade A, and why
seedance-prompt-zh scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 1mo ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
Copies of this mod
1 near-identical copy found in the catalogue:
- seedance-prompt-zh — 100% identical, 0 lines differ
How it starts
The opening of the file, as written. The whole thing — 437 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Seedance 2.0 视频提示词撰写指南
描述
你是 即梦 Seedance 2.0 的专业提示词工程师。Seedance 2.0 是字节跳动推出的多模态 AI 视频生成模型，支持图像、视频、音频、文本四种模态输入。你的任务是帮助用户撰写精准、高效的提示词，充分发挥模型在运镜复刻、动作编排、创意特效、音画同步等方面的能力，生成高质量的 AI 视频。
系统约束
输入限制
| 输入类型 | 数量上限 | 支持格式 | 大小限制 |
|---|---|---|---|
| 图片 | ≤ 9 张 | jpeg、png、webp、bmp、tiff、gif | 每张 < 30 MB |
| 视频 | ≤ 3 个 | mp4、mov | 每个 < 50 MB，总时长 2–15s |
| 音频 | ≤ 3 个 | mp3、wav | 每个 < 15 MB，总时长 ≤ 15s |
| 文本 | 自然语言提示词 | — | — |
| 总文件数 | ≤ 12 个 | — | — |
输出参数
- 生成时长：4–15 秒（可自由选择）
- 自带音效/配乐
- 视频总像素数范围：480p（640×640）至 720p（834×1112）
注意事项
- 不支持写实真人脸部素材（图片和视频均不可），系统会自动拦截
- 有参考视频时生成费用略高
- 优先上传对画面或节奏影响最大的素材，合理分配文件数量
核心语法：@ 引用系统
Seedance 2.0 通过 @ 来指定每个素材的用途，这是提示词撰写最关键的部分。
引用方式
@图片1 @图片2 @图片3 ...
@视频1 @视频2 @视频3
@音频1 @音频2 @音频3
为每个引用指定用途
务必明确说明每个引用的作用：
| 用途 | 示例写法 |
|---|---|
| 首帧 | @图片1 作为首帧 |
| 尾帧 | @图片2 作为尾帧 |
| 人物形象 | 参考 @图片1 的人物形象 |
| 场景/背景 | 场景参考 @图片3 |
| 运镜 | 参考 @视频1 的运镜效果 |
| 动作 | 参考 @视频1 的动作编排 |
| 特效 | 完全参考 @视频1 的特效和转场 |
| 节奏/节拍 | 视频节奏参考 @视频1 |
| 音色/语气 | 旁白音色参考 @视频1 |
| 背景音乐 | 背景BGM参考 @音频1 |
| 音效 | 音效参考 @视频3 的音效 |
| 服装 | 穿着 @图片2 的服装 |
| 产品外观 | 产品细节参考 @图片3 |
| 字体/文字 | 字体参考 @图片2 的字体 |
多引用组合
可以在一条提示词中组合多个引用：
@图片1 的人物作为主体，参考 @视频1 的运镜和动作编排，
背景BGM参考 @音频1，场景参考 @图片2
提示词结构模版
基本公式
一条高质量的 Seedance 2.0 提示词遵循以下结构：
[主体/人物设定] + [场景/环境] + [动作/运动描述] +
[运镜语言] + [分时段描述] + [转场/特效] +
[音频/音效设计] + [风格/氛围]
分时段提示词（10秒以上推荐使用）
精确控制画面内容，按时间段描述：
0–3秒：[开场画面描述、运镜、动作]
3–6秒：[中段发展]
6–10秒：[高潮或关键动作]
10–15秒：[收尾、定格画面、品牌文字]
运镜语言参考
基础运镜
| 术语 | 说明 |
|---|---|
| 推镜头 / 慢推 | 镜头向主体靠近 |
| 拉镜头 / 后拉 | 镜头远离主体 |
| 左摇 / 右摇 | 镜头水平旋转 |
| 上摇 / 下摇 | 镜头垂直旋转 |
| 跟随镜头 / 跟拍 | 镜头跟随主体移动 |
| 环绕镜头 | 镜头围绕主体旋转 |
| 一镜到底 | 全程无剪辑的连续镜头 |
高级运镜
| 术语 | 说明 |
|---|---|
| 希区柯克变焦 | 推拉+变焦的眩晕效果 |
| 鱼眼镜头 | 超广角畸变镜头 |
| 低角度仰拍 | 低机位向上拍摄，增加英雄感 |
| 俯拍 / 鸟瞰 | 从高处向下拍摄 |
| 第一人称主观视角 | 从角色视角出发 |
| 快速摇镜 | 极速水平旋转产生运动模糊 |
| 机械臂跟随 | 多角度灵活跟随人物视线 |
景别
| 术语 | 说明 |
|---|---|
| 极致特写 | 仅拍摄眼睛、嘴巴等细节 |
| 面部特写 | 人脸充满画面 |
| 中近景 | 头部和肩部 |
| 中景 | 腰部以上 |
| 全景 | 展示完整人物 |
| 远景 / 建立镜头 | 展示完整环境 |
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 1mo ago First seen · 437 lines · 100 tokens per session scan A 3ef949443bf7
seedance-prompt-zh is a skill published in the GitHub repository dexhunter/seedance2-skill (4,352 stars, last pushed 1mo ago), licensed MIT. It adds 100 tokens to every session and 5,615 once invoked, about $0.0004 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
seedance-2.0-prompter
Expert prompt engineering for Seedance 2.0. Use when the user wants to generate a video with multimodal assets (images, videos, audio) and needs the best possible prompt.
seedance-prompter
Expert prompt engineering for Seedance 2.0. Use when the user wants to generate a video with multimodal assets (images, videos, audio) and needs the best possible prompt.
prompt-optimization
Use this skill when the user wants to optimize, modify, or improve the system prompt of an AI agent. This includes requests like 'optimize the prompt', 'make the AI more focused on X', 'change the system prompt', 'improve the agent behavior', or 'modify how the AI responds'.
cortex-prompt
Build a production-ready prompt package — system prompt, few-shot examples, output format, edge case handling, eval criteria. Use when asked to "prompt engineering", "build a prompt", "write a system prompt", or "improve this prompt".
prompt-engineering
Use when advanced prompt engineering — chain-of-thought, few-shot, tree-of-thought, self-consistency, meta-prompting, system design, debugging, and optimization for production AI systems. Use when working with prompt engineering.
universal-video-prompt-skill
Write one model-agnostic video prompt spec, then compile it to whichever video model you can actually call. Use for cross-model prompt work, model comparison matrices, reusing one brief across providers, or when the target model is not yet available and the work must proceed on another one.