Agent skill

Seedance Prompt Review Skill

by Supreme-Ultimate in Supreme-Ultimate/novel-to-script-team

Seedance 提示词审核技能。用于审核阶段三分镜师产出的 Seedance 2.0 动态提示词,通过逐条比对、脑内预演和评分制确保规范性、运镜合理性、叙事连贯性。

MITAuto-check passedMedia & Creative

Install Seedance Prompt Review Skill

skills CLI
$ npx skills add Supreme-Ultimate/novel-to-script-team --skill seedance-prompt-review-skill -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install Supreme-Ultimate/novel-to-script-team seedance-prompt-review-skill --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/Supreme-Ultimate/novel-to-script-team.git skills-src && mkdir -p .claude/skills && cp -r skills-src/skills/seedance-prompt-review-skill .claude/skills/seedance-prompt-review-skill && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
seedance-prompt-review-skill
GitHub stars
175
Token cost
~1.3k tokens
SKILL.md length
250 words
Files
1
Skills in repo
17
Repo updated
First seen
Licence
MIT

At a glance

Seedance 提示词审核技能。用于审核阶段三分镜师产出的 Seedance 2.0 动态提示词,通过逐条比对、脑内预演和评分制确保规范性、运镜合理性、叙事连贯性。

  • Works in 4 steps: ../../references/00-first-principles.md → ../../references/04-review-gates.md → ../../references/05-compliance-boundaries… → …
  • Tasks that involve AI video generation
  • SKILL.md covers 必读, 技能说明, 审核对象 and 审核流程, plus 3 more sections
  • Instructions only: no scripts, shell commands, URLs or credentials in SKILL.md

What it does

Seedance Prompt Review Skill is an agent skill from Supreme-Ultimate/novel-to-script-team. Seedance 提示词审核技能。用于审核阶段三分镜师产出的 Seedance 2.0 动态提示词,通过逐条比对、脑内预演和评分制确保规范性、运镜合理性、叙事连贯性。

Its SKILL.md is about 1.3k tokens, which your agent loads only when the skill is triggered. It is a single SKILL.md file with no bundled scripts.

It sits in Media & Creative, covering AI video generation. It works with Seedance. The repository describes itself as: 完整的多 Agent 多 Skill 小说改编影视流水线系统。 The licence is MIT.

When your agent uses it

  • Tasks that involve AI video generation

Example prompts

  • “/seedance-prompt-review-skill”

Workflow steps

4 steps, taken from the first numbered list in SKILL.md.

  1. ../../references/00-first-principles.md
  2. ../../references/04-review-gates.md
  3. ../../references/05-compliance-boundaries.md
  4. ../../references/20-frame-description-elements.md(帧图描述完整元素表)

What it can do on your machine

Read from SKILL.md and the folder at commit 117dceb. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    No scripts in the folder and no shell commands in SKILL.md.

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    No URLs in SKILL.md.

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Seedance Prompt Review Skill loads about 1.3k tokens when it runs. Until then it costs about 28 tokens; SKILL.md has 250 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~28
When it runs · the whole SKILL.md, loaded when a task matches
~1.3k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from Supreme-Ultimate/novel-to-script-team at commit 117dceb, republished under its MIT licence (© Supreme-Ultimate). 250 words, ~1,272 tokens.

Download SKILL.mdSave it as .claude/skills/seedance-prompt-review-skill/SKILL.md (or your agent's skills folder).
name
seedance-prompt-review-skill
description
Seedance 提示词审核技能。用于审核阶段三分镜师产出的 Seedance 2.0 动态提示词,通过逐条比对、脑内预演和评分制确保规范性、运镜合理性、叙事连贯性。

Seedance 2.0 提示词审核技能

必读

  1. ../../references/00-first-principles.md
  2. ../../references/04-review-gates.md
  3. ../../references/05-compliance-boundaries.md
  4. ../../references/20-frame-description-elements.md(帧图描述完整元素表)

技能说明

审核分镜师在阶段三的 Seedance 2.0 提示词产出。导演需要以批判性视角审视分镜师的产出,通过逐条比对导演讲戏本、脑内预演 Seedance 生成效果、逐条评分的方式,确保提示词质量真正达标,能通过 Seedance 2.0 生成符合导演意图的视频。

审核对象

outputs/<集数>/02-seedance-prompts.md

审核流程

第一步:建立整体理解
  • 读取导演讲戏本(01-director-analysis.md)
  • 读取人物提示词(outputs/{剧本名}/assets/character-prompts.md)和场景道具提示词(outputs/{剧本名}/assets/scene-prompts.md)
  • 读取待审核的 Seedance 提示词产出
  • 明确审核立场:你的任务是找出问题,而非确认通过
第二步:逐条比对
  • 对每条提示词,逐句对照导演讲戏本中对应剧情点的导演阐述
  • 找出导演阐述中有但提示词中被弱化、遗漏或表达偏离的地方
  • 找出分镜师自行添加的、导演阐述中没有的内容,判断是否合理
  • 记录所有偏差,无论大小
第三步:脑内预演
  • 对每条提示词,想象它被送进 Seedance 2.0 后,生成的视频画面是什么样的
  • 判断:生成的视频是否与导演讲戏本中描述的画面一致?
  • 判断:是否有描述方式可能导致 Seedance 理解错误(如动作顺序混乱、运镜方向矛盾、多个动作在短时间内无法完成)?
  • 判断:@引用的参考图加上提示词的动态描述,生成效果是否会自然?
第四步:逐条评分

按照下方验收清单,对每条提示词评分(1-10)

第五步:输出结果
  • 计算所有提示词的平均分
  • 平均分 ≥ 8 且无单项低于 6 → PASS
  • 平均分 < 8 或任一单项低于 6 → FAIL + 问题清单

验收清单

生成模式选型检查(整体检查,不评分)
  • □ 模式合理性:硬切换景别/机位的镜头是否使用了 I2V-首帧 或 I2V-首帧 + 多参考?连续动作过渡的镜头是否使用了 首尾帧?纯氛围/概念镜头是否使用了 T2V?逐条检查,列出不合理的模式选择。
  • □ 帧图需求完整性:所有 I2V-首帧 模式的提示词是否都有对应的首帧图需求?所有 首尾帧 模式的提示词是否都有首帧图和尾帧图需求?逐条检查,列出遗漏。
  • □ 帧图描述质量:帧图描述是否为精确的静态单帧画面(构图、人物姿态、光影、色调)?是否包含动态描述(帧图不应有动作)?逐条检查。
  • □ 引用配额检查:叠加模式(如 I2V-首帧 + 多参考)下,帧图 + 参考图总数是否 ≤ 9 张?逐条检查。
  • □ 模式标注格式:每条提示词是否都标注了 **生成模式** 和 **选型理由**?格式是否统一?
素材对应表(整体检查,不评分)
  • □ 完整性:是否为所有引用的素材建立了对应关系?逐个检查,列出遗漏。
  • □ 准确性:素材描述是否与 character-prompts.md / scene-prompts.md 一致?逐个比对,列出不一致。
  • □ 无孤立引用:提示词中的 @引用是否都在对应表中有记录?逐条检查。
  • □ 素材边界正确:对应表中是否只包含在 character-prompts.md / scene-prompts.md 中有参考图的素材?不在素材文件中的群演/一次性配角不应出现在对应表中。
  • □ 场景独立编号:scene-prompts.md 中的九宫格是生成格式,用户会将每个格子拆成独立的场景参考图。对应表中每个场景是否独立编号为一个 @图片?如果发现整张九宫格被作为一个 @图片(如"@图片5 = ep01 场景宫格"),则为严重错误,必须 FAIL。
  • □ 引用用途明确:每个 @引用是否都明确标注了用途(如"参考@图片1的人物形象")?是否存在模糊引用(如只写"参考@视频1"而不说参考什么)?逐个检查。
  • □ 平台约束合规(单条维度):每条提示词内的素材引用是否在 Seedance 2.0 单次生成限制内?图片 ≤ 9、视频 ≤ 3、音频 ≤ 3、总文件 ≤ 12。注意:素材对应表是全文档总映射,@图片总数可以超过 9 张。
群演/一次性配角处理(整体检查,不评分)
  • □ 导演讲戏本中提到但不在 character-prompts.md 中的人物(群演、一次性配角),在提示词中是否用文字直接描述其外观?
  • □ 这些人物是否错误地使用了 @引用?(不应使用)
  • □ 文字描述是否足够具体,能让 Seedance 生成符合导演意图的人物形象?
帧图元素完整性检查(整体检查,不计分)

对 03-frame-requests.md 中每个帧图描述进行元素完整性检查:

检查项标准
所有帧图必备 7 项景别、机位角度、方位构图、人物位置、人物朝向、姿态动作、光线类型
近景/特写额外 3 项表情眼神、光源方向、光影效果
全景/远景额外 4 项前景元素、背景环境、景深效果、人物与环境距离
多人场景额外 2 项人物关系、人物与人物关系
相对位置描述人物与环境距离、物品与空间关系是否清晰

输出格式:

帧图景别必备 7 项额外必备缺失元素判定
F01中景7/73/3无✅
F02特写7/72/3光影效果⚠️

缺失率超过 20% 的帧图必须标记为 FAIL 并要求补全。

单条提示词质量(每条提示词单独评分 1-10)

评分维度:

  • □ 忠实度(逐句比对):将导演阐述和提示词逐句对照,核心动作、台词、镜头运动、光影描述是否被完整保留?哪些被弱化或遗漏了?哪些被偏离了?具体列出。
  • □ 画面还原度(脑内预演):想象 Seedance 根据这条提示词生成的视频,画面是否与导演讲戏本中描述的画面一致?如果有偏差,具体哪个画面元素会出问题?
  • □ 动作可执行性(脑内预演):每个连续镜头内的动作密度是否合理(1 拍 ≈ 2.5 秒)?多镜头段落的节拍是否合理分配在各镜头中?是否存在动作跳跃?具体指出。
  • □ 镜头可实现性(脑内预演):单镜头段落——镜头运动是否连续可实现?多镜头段落——每组镜头各自是否可执行,切换方式是否清晰?是否存在物理上矛盾的要求?具体指出。
  • □ Seedance 友好度:提示词是否采用叙事描述式?是否有否定句?是否有重复静态内容?是否有可能让 Seedance 困惑的歧义表达?
  • □ 音频设计:提示词中是否包含音频指导?
    • 标准版本(≤60秒):应包含背景音乐风格、环境音效、动作音效、台词语气
    • 静音版本(>60秒超长视频):可仅保留环境音和对白,明确标注"不添加背景音乐"
    • 音频描述是否与画面情绪匹配?
    • 如果有台词/对白,是否标注了语气?
    • 如果完全缺失音频设计(连环境音都没有),扣分
  • □ 情绪准确性:这条提示词传达的情绪基调是否与导演讲戏本中该剧情点的核心情绪一致?如果有偏差,指出导演要的情绪是什么,提示词传达的变成了什么。

评分标准:

  • 9-10:提示词精准,脑内预演生成视频完全符合导演讲戏意图,无需修改
  • 7-8:基本到位,但有 1-2 处可能导致生成偏差的地方
  • 5-6:有明显的遗漏、偏离或可执行性问题,需要修改
  • 1-4:提示词严重偏离导演意图或无法生成有效视频
整体连贯性(整体评估,不单独评分,但影响最终判定)
  • □ 衔接自然:按顺序读完所有提示词,相邻两条之间的画面空间、时间、情绪是否能自然过渡?如果不能,指出具体哪两条之间有断裂。
  • □ 节奏合理:整体节奏是否有起伏变化?是否存在连续多条节奏相同导致观感疲劳的情况?
  • □ 情绪弧线:全部提示词串联起来,情绪弧线是否与导演的叙事设计一致?如果有偏差,指出哪里开始偏离。
  • □ 风格统一:全篇提示词的写法风格、详略程度是否一致?是否有某条突然变得过于简略或过于冗长?
  • □ 光影连续:相邻场景之间的光影/色调过渡是否合理?是否与导演设定的光影变化一致?

输出规范

评分表格式
| 提示词 | 忠实度 | 画面还原 | 动作可执行 | 镜头可实现 | Seedance友好 | 音频设计 | 情绪准确 | 均分 | 问题摘要 |
|--------|--------|---------|-----------|-----------|-------------|---------|---------|------|---------|
| P01    | 9      | 9       | 8         | 9         | 9           | 8       | 9       | 8.7  | 无 |
| P02    | 8      | 7       | -         | 8         | 9           | 7       | 7       | 7.7  | 缺少环境音效描述 |
| ...    |        |         |           |           |             |         |         |      | |

**平均分:X.X/10**

整体连贯性评估:[简述衔接、节奏、情绪弧线的判断]
PASS 输出(平均分 ≥ 8 且无单项低于 6)
✅ **业务审核:PASS**

[评分表]

整体评价:[1-2 句总结]
FAIL 输出(平均分 < 8 或任一单项低于 6)
❌ **业务审核:FAIL**

[评分表]

**需要修改的问题**:

**问题1**:[具体问题描述]
- 位置:[如 P03 提示词]
- 当前问题:[具体描述现在的内容哪里不对]
- 脑内预演:[想象 Seedance 生成效果会出什么问题]
- 修改方向:[具体建议怎么改]

**问题2**:...

---
**修改优先级**:
1. 🔴 必须修改(单项低于 6 的):[列出]
2. 🟡 建议修改(单项 6-7 的):[列出]

成功标准

  1. 所有提示词平均分 ≥ 8
  2. 无单项评分低于 6
  3. 素材对应表完整准确,场景独立编号
  4. 群演/一次性配角处理正确
  5. 整体连贯性良好
  6. 问题清单具体明确,包含位置、问题、脑内预演、修改方向
  7. 修改优先级清晰标注
  8. 生成模式选型合理(硬切用 I2V、连续动作用首尾帧、纯氛围用 T2V)
  9. 帧图需求完整(I2V/首尾帧模式都有对应帧图)
  10. 叠加模式下引用配额 ≤ 9 张

© Supreme-Ultimate, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

Just SKILL.md in skills/seedance-prompt-review-skill of Supreme-Ultimate/novel-to-script-team.

Open the folder on GitHubat commit 117dceb

Compare with similar skills

Seedance Prompt Review Skill next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Seedance Prompt Review Skill compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Seedance Prompt Review Skill this skillSupreme-Ultimate/novel-to-script-team175—~1.3kAutomated safety check: PassMIT
Seedancesongguoxs/seedance-prompt-skill2.9k1 repos~2.5kAutomated safety check: PassNone
SN Motion HTMLOpenSenseNova/SenseNova-Skills5.7k—~2.2kAutomated safety check: NotesMIT
Seedance Prompt Endexhunter/seedance2-skill4.2k—~3.8kAutomated safety check: PassMIT
Seedance Japanese Prompt ExamplesEmily2040/seedance-2.07.6k1 repos~898Automated safety check: PassMIT
Seedance 2 5 Video Directorliyue-aigc/seedance-2-5-video-director599—~3.5kAutomated safety check: PassMIT

Similar skills

  • Seedance

    songguoxs/seedance-prompt-skill

    This skill should be used when the user asks to "generate video prompts", "create Seedance prompts", "write video descriptions", mentions "Seedance", "seedance", "即梦", "即梦平台", "视频提示词", "视频生成"…

    2.9k GitHub starsUsed in 1 repo~2.5k tokens
    Media & CreativeAuto-check passed
  • SN Motion HTML

    OpenSenseNova/SenseNova-Skills

    Builds HTML stories where one continuous camera journey advances with page progress, using researched structure, AI stills, Seedance video clips and browser QA.

    5.7k GitHub stars~2.2k tokensUpdated 2 days ago
    Media & CreativeAuto-check: notes
  • Seedance Prompt En

    dexhunter/seedance2-skill

    Write effective prompts for Jimeng Seedance 2.0 multimodal AI video generation.

    4.2k GitHub stars~3.8k tokensUpdated 21 days ago
    Media & CreativeAuto-check passed
  • Seedance Japanese Prompt Examples

    Emily2040/seedance-2.0

    Provides Japanese prompt patterns and safe example rewrites for Seedance 2.0 video generation, with reference tags kept intact and each example labeled by risk.

    7.6k GitHub starsUsed in 1 repo~898 tokens
    Media & CreativeAuto-check passed
  • Seedance 2 5 Video Director

    liyue-aigc/seedance-2-5-video-director

    Design, derive, optimize, diagnose, and rewrite scripts or copy-ready prompts specifically for Dreamina/即梦 Seedance 2.5.

    599 GitHub stars~3.5k tokensUpdated 1 mo ago
    Media & CreativeAuto-check passed
  • Seedance Prompt Zh

    dexhunter/seedance2-skill

    为即梦 Seedance 2.0 多模态AI视频生成模型撰写高质量提示词。当用户需要使用文本、图片、视频、音频等多模态输入创作视频提示词时触发。涵盖@引用语法、运镜复刻、特效模仿、视频延长、视频编辑、音乐卡点、电商广告、短剧创作、科普教育等场景。

    4.2k GitHub stars~1.9k tokensUpdated 21 days ago
    Media & CreativeAuto-check passed

More from Supreme-Ultimate/novel-to-script-team

All 17 skills in this repo
  • Production Art Prompt Designer

    Supreme-Ultimate/novel-to-script-team

    Turns the character and scene lists from a director analysis into detailed text-to-image prompts for character sheets, scenes and props, after confirming a visual style with you.

    175 GitHub stars~1.8k tokensUpdated 5 mo ago
    Auto-check passed
  • Art Direction Review

    Supreme-Ultimate/novel-to-script-team

    Reviews costume, makeup and prop design prompts for characters and scenes in a novel-to-film pipeline, scoring each one and returning PASS or FAIL with a problem list.

    175 GitHub stars~918 tokensUpdated 5 mo ago
    Auto-check passed
  • Script Comparative Review Against Hits

    Supreme-Ultimate/novel-to-script-team

    Chinese-language skill that scores a generated short-drama script against a reference hit script on rhythm, dialogue style, emotional impact and structure, with PASS/FAIL thresholds.

    175 GitHub stars~1.8k tokensUpdated 5 mo ago
    Auto-check passed
  • Director Script Breakdown for Seedance

    Supreme-Ultimate/novel-to-script-team

    Breaks a script or outline into plot points and writes a director's walkthrough, character list and scene list sized for Seedance video generation.

    175 GitHub stars~1.3k tokensUpdated 5 mo ago
    Auto-check passed
  • Film Storyboard Prompt Builder

    Supreme-Ultimate/novel-to-script-team

    Turns a script, synopsis or scene outline into AI image prompts for storyboards, using a beat breakdown, a nine-panel Beat Board and four-panel Sequence Boards.

    175 GitHub stars~815 tokensUpdated 5 mo ago
    Auto-check passed
  • Hit Script Retrieval Skill

    Supreme-Ultimate/novel-to-script-team

    爆款剧本检索技能。使用混合搜索(语义+关键词)检索最相关的爆款剧本,为生成提供参考. An agent skill from Supreme-Ultimate/novel-to-script-team.

    175 GitHub stars~1.6k tokensUpdated 5 mo ago
    Auto-check passed

Works with

Questions about Seedance Prompt Review Skill

What does Seedance Prompt Review Skill do?

Seedance 提示词审核技能。用于审核阶段三分镜师产出的 Seedance 2.0 动态提示词,通过逐条比对、脑内预演和评分制确保规范性、运镜合理性、叙事连贯性。. Seedance Prompt Review Skill is an agent skill from Supreme-Ultimate/novel-to-script-team.

When should I use Seedance Prompt Review Skill?

Seedance Prompt Review Skill fits situations like: tasks that involve AI video generation.

How do I install Seedance Prompt Review Skill in Claude Code?

Run `npx skills add Supreme-Ultimate/novel-to-script-team --skill seedance-prompt-review-skill -a claude-code`. Or copy the skill folder (skills/seedance-prompt-review-skill in Supreme-Ultimate/novel-to-script-team) into .claude/skills/seedance-prompt-review-skill in your project. Claude Code loads it when a task matches its description.

How do I install Seedance Prompt Review Skill in Codex?

Run `npx skills add Supreme-Ultimate/novel-to-script-team --skill seedance-prompt-review-skill -a codex`. Or copy the skill folder (skills/seedance-prompt-review-skill in Supreme-Ultimate/novel-to-script-team) into .agents/skills/seedance-prompt-review-skill in your project. Codex loads it when a task matches its description.

Can I use Seedance Prompt Review Skill in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add Supreme-Ultimate/novel-to-script-team --skill seedance-prompt-review-skill -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/seedance-prompt-review-skill, .gemini/skills/seedance-prompt-review-skill, .github/skills/seedance-prompt-review-skill and .opencode/skills/seedance-prompt-review-skill in your project.

What does Seedance Prompt Review Skill need to run?

SKILL.md names no scripts, command-line tools or credentials: Seedance Prompt Review Skill is instructions for the agent only.

Does Seedance Prompt Review Skill access the network?

SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.

Is Seedance Prompt Review Skill safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.

What licence does Seedance Prompt Review Skill use?

Seedance Prompt Review Skill is published under the MIT licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Seedance Prompt Review Skill use?

About 1.3k tokens (SKILL.md is roughly 5.1k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.

What are the alternatives to Seedance Prompt Review Skill?

Skills that share tags, products or a category with Seedance Prompt Review Skill: Seedance (songguoxs/seedance-prompt-skill, 2.9k stars), SN Motion HTML (OpenSenseNova/SenseNova-Skills, 5.7k stars), Seedance Prompt En (dexhunter/seedance2-skill, 4.2k stars) and Seedance Japanese Prompt Examples (Emily2040/seedance-2.0, 7.6k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Seedance Prompt Review Skill?

Supreme-Ultimate (a GitHub organization) maintains it in Supreme-Ultimate/novel-to-script-team, which has 175 GitHub stars. The repository holds 17 skills in this directory. The repository was last updated on May 2, 2026.

Source: Supreme-Ultimate/novel-to-script-team on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.