Skip to content
Back to search

lengyi-seedance2.5-prompt

设计 Seedance 2.5 视频提示词。输入一段文本(小说片段、一句剧情或几个主题词),扩写成 20-30s(最长

Safety scan

A static text scan, not a guarantee. Select the badge to see what was scanned and every finding.

Checked today
Badge for your README
findsafeskills safety rating

Markdown

[![findsafeskills safety rating](/api/listings/woyin2024--lengyi-seedance2-5-prompt/badge.svg)](/listings/woyin2024--lengyi-seedance2-5-prompt)

HTML

<a href="/listings/woyin2024--lengyi-seedance2-5-prompt"><img src="/api/listings/woyin2024--lengyi-seedance2-5-prompt/badge.svg" alt="findsafeskills safety rating" /></a>

Install

i

Install instructions from the README

As written by the repo's author. findsafeskills has not verified these steps.

Who Is This Skill For

Any AI Agent that supports the `Agent Skills` spec (e.g. Codex, Claude Code, WorkBuddy, TRAE Work, Qoder Work, Kimi Work, etc.). As long as your Agent can read and obey a Skill, it can use this to design Seedance 2.5 video prompts.

In principle, models with stronger reasoning are recommended, e.g. Claude Opus 5, K3, DeepSeek-v4-pro, Qwen3.8, GPT-5.6, etc.

Installation

# Clone the repo
git clone https://github.com/woyin2024/lengyi-seedance2.5-prompt.git

Drop the entire `lengyi-seedance2.5-prompt` folder into your Agent's Skills directory:

- **Claude Code project-level**: `<project>/.claude/skills/lengyi-seedance2.5-prompt/` - **WorkBuddy user-level**: `~/.workbuddy/skills/lengyi-seedance2.5-prompt/` - **Other Agents**: follow your tool's Skill installation guide

Usage

Once installed, trigger it in conversation any of these ways:

- "Turn this novel excerpt into a Seedance video storyboard" - "Use the lengyi-seedance2.5-prompt skill to design a 30-second multi-shot prompt, the theme is…" - "Make a shotlist, image-to-video"

Trigger keywords (Chinese): `seedance 视频提示词`, `多镜头提示词`, `视频分镜脚本`, `镜头脚本`, `shotlist`, `文生视频 / 图生视频`, `AI 视频镜头脚本`.

Recommended Workflow

1. Use the Skill to generate the "Asset Image T2I Prompts" first
2. Feed the T2I prompts into a text-to-image model (e.g. Image2) to generate
   subject / scene asset images
3. Feed the asset images + the Skill's "Video Prompt" output into Seedance 2.5

---

What was scanned

The files the safety scan read, as written by the repository's author and shown as plain text. Lines with a finding are marked.

SKILL.md · Manifest · 154 lines

View this file on GitHub (read at the repository's current version: the scan did not record which commit it read)

  1. ---
  2. name: lengyi-seedance2.5-prompt
  3. description: 设计 Seedance 2.5 视频提示词。输入一段文本(小说片段、一句剧情或几个主题词),扩写成 20-30s(最长
  4. 30s)的连续镜头脚本,输出一条统一格式的多镜头提示词(主体 / 场景 / 音乐 +
  5. 镜头1、镜头2…),镜头数根据故事和节奏合理安排。先锚定主体与场景资产图,给出对应的文生图提示词,以锁定一致性。当用户要做 seedance
  6. 视频提示词、多镜头提示词、视频分镜脚本、镜头脚本、shotlist、文生视频/图生视频、AI 视频镜头脚本时使用。
  7. ---
  8. # lengyi-seedance2.5-prompt skill:Seedance 2.5 视频提示词设计
  9. 把用户输入的文本设计成 **20-30s 的连续镜头脚本**,并输出为**一条**可直接复制进 Seedance 2.5 模型的多镜头视频提示词:头部锚定主体、场景、音乐,正文按 `镜头N | …` 顺序排列。
  10. ## 默认前提(硬性)
  11. - **必须先锚定资产**:主体与场景都是视频的"资产图",每次输出都要先设计出它们的**文生图(T2I)提示词**,再写视频提示词(见第 4 步)。头部 `主体:`/`场景:` 里填的是这些资产的名称。
  12. - **音乐行固定不变**:`音乐:不要生成任何背景音乐,只生成对应音效`。这一行是全片唯一的声音指令——**各镜头里不再描述任何音效**,音效交由模型根据画面自动匹配。
  13. - **镜头正文只描述画面**:不写音效,不写剧情作用/叙事目的。叙事在设计阶段想清楚,但不落进提示词。
  14. - 默认用中文写提示词。用户要求或目标为海外工具时给英文/中英双语。
  15. ## 能力边界
  16. - 输入三类,分别处理:①小说/剧情/脚本片段→提炼最具画面感的故事节奏;②一句简短剧情→补全前因后果与转折;③几个主题词→创意扩写出"起承转合"短片创意,再拆镜头。
  17. - 输出必须明确:​**总时长(默认 20-30s,上限 30s)**、**镜头数(按内容定,无固定档位)**、每镜时间区间。
  18. - 用户显式指定时长/镜头数则优先;超过 30s 时按 30s 处理并一句话说明。
  19. ## 工作流程(严格按序执行)
  20. ### 第 1 步 · 解析与扩写输入
  21. 在心里列出并补全:
  22. 1. **故事节奏:​**拆成有先后、连续性、逻辑紧密的故事画面。故事节奏是你的设计依据,写进【脚本总览】即可,不写进视频提示词。
  23. 2. **主体**:有几个角色/物体/道具主体?各自外观线索(缺失则合理设定)。
  24. 3. **场景**:地点、时间、天气、空间。跨场景时列清场景数量与切换点。
  25. 4. **基调/类型**:从故事气质推断(科幻/悬疑/温情/热血/广告/武侠…)。
  26. ### 第 2 步 · 定总时长与镜头数(由故事和节奏决定)
  27. **镜头数完全取决于故事和节奏**,没有固定档位,也没有数量上限。流程是:定总时长 → 排叙事节拍 → 给每个节拍选镜头类型 → 按类型给时长 → 加总校准。​**不要用"总时长 ÷ 平均值"倒推镜头数。,**
  28. - 总时长:默认 **20-30s**;内容简单取 20-24s,叙事完整取 25-30s。用户另有要求时,则听用户。,
  29. - **每镜时长按镜头类型定**(核心依据):
  30. | 镜头类型 | 时长 |
  31. | - | - |
  32. | 特写 / 快节奏镜头 | 2-3s |
  33. | 中景 / 叙事镜头 | 3-6s |
  34. | 全景 / 固定镜头 | 6-7s |
  35. - **镜头数是上面选择的结果**。参考量级:30s 视频通常落在 **8-14 镜**(均值约 2-4s);20s 约 7-10 镜。全景、固定镜头多则镜头数偏少,特写快切多则偏多。镜头的数量主要取决于**故事**和**节奏**。
  36. **反狂切镜头规则**(约束的是"无理由地切",不是"短镜头"):
  37. 1. **每一次切镜都要有叙事理由**——换空间、换时间、换视角、换信息层级、情绪断点。为炫技而切、为"看起来热闹"而切,一律合并。这是唯一真正的反狂切标准。
  38. 2. **能一镜说完的不切两镜**:优先用「运镜 + 主体动作 + 环境变化」在一镜内完成推进(推近代替切特写、跟拍代替切两个机位)。
  39. 3. **分镜节奏要有起伏,不能通篇同一时长**:全篇都是 2-3s 就是狂切,全篇都是 6s 则拖沓。长短交替,让快剪段落成为对比出来的高潮。
  40. 4. **快剪成组使用**:连续 2-3s 的短镜适合表现打斗、蒙太奇、情绪爆点,成组出现后要用一个中景或全景把节奏收住。
  41. 5. 长镜头是优点:建立镜可用 6-7s,靠运镜与动作分层填满,不必急着切。
  42. **时长算术(唯一硬约束)**:各镜时长之和 = 总时长,且总时长 ≤ 30s。每镜标明时间区间(如 `0.0–6.5s`),自检末镜结束时间 = 总时长。
  43. **节奏规则**:开场用全景/建立镜(6-7s)交代环境与主体 → 中段用中景叙事镜稳步推进 → 高潮切入 2-3s 特写快剪组 → 结尾回到较长镜头收束(≥3s)。
  44. ### 第 3 步 · 建立主体与场景档案(一致性核心)
  45. 1. 为**每个主体**写**锚点描述**——固定不变的视觉特征:年龄、发型发色、瞳色、脸型、服装/材质、标志性道具。措辞一旦确定,全程不改。
  46. 2. 为**核心场景**写场景锚点(地点、时间、光线色调、天气、空间布局);同场景全程沿用同一措辞。
  47. 3. 给每个主体/场景取一个**短名(资产名)**,如「武松」「猛虎」「景阳冈山道」。头部 `主体:`/`场景:` 只填这些短名,镜头正文也只用短名。
  48. 4. **多主体**:列出人物关系与出场顺序;每镜标注在场主体。
  49. 5. **风格**:全片一句统一的风格关键词,写在资产锚定区块,视频提示词里只在镜头1点明一次,后续不重复。
  50. 6. 锚点档案是**资产图生成依据**,**不在各镜头里逐字复述**(原因见第 5 步)。
  51. ### 第 4 步 · 写资产图文生图(T2I)提示词(每次必做)
  52. 主体和场景都要给 T2I 提示词,先生成资产图,再连同视频提示词一起使用:
  53. - **主体资产(人物/物体/道具)**:背景必须简洁(纯色/白底/浅景深虚化),​**只突出主体本身**,不带叙事场景,便于模型提取特征。内容 = 逐字复述锚点描述 + 姿态/角度(正面视角、全身或半身清晰)+ 统一风格 + 画质词(8K、精致细节、studio 柔光)。
  54. - **场景资产**:展示核心场景的**代表性全貌**,**排除人物与临时道具**,光线/色调/布局与场景锚点一致 + 统一风格 + 画质词。
  55. - **数量**:主体按实际数量给(通常 1-3 个),场景按实际场景数给(通常 1-2 个)。保持精简,不要为凑数编造资产。
  56. ### 第 5 步 · 写多镜头视频提示词(只写画面,且是会动的画面)
  57. **两条核心原则:、**
  58. - **只描述画面**:镜头正文写的是"摄影机拍到了什么、怎么动的",重点是介绍视频的画面内容。不写音效,不写"剧情:本镜推进了什么",不写情绪解释与叙事意图。
  59. - **是视频不是静帧**:每镜是一段会动的视频片段,禁止写成静态画面 caption。
  60. 严格按统一模板(见"输出格式")。每个 `镜头N` 一行,必须含以下要素:
  61. - **时间区间**(写在镜号后)
  62. - **场景短名**(多场景时每镜标明当前场景;单场景可在镜头1标明后省略)
  63. - **主体**:**仅用短名(如"武松")+ 本镜变化点**(如"毡笠帽背系脊梁、发束被风吹乱")。​**不复述固定外观**——外观由资产图锁定,重复描写会抢占模型注意力、弱化动作与运镜。
  64. - **景别 + 镜头运动**:明示方向与速度(缓慢推近 / 拉远 / 上摇 / 横移 / 跟拍 / 环绕 / 手持轻晃 / 慢动作 / POV)。景别要与第 2 步定的时长匹配。6-7s 的建立镜必须写出运镜的**阶段变化**(如"先低角度缓推,至中段上摇至全景");2-3s 的特写只给一个明确、干净的运镜,别塞复杂运动。
  65. - **动作过程**:强动词写出本镜内的连续动作变化("步伐由快转慢→缓缓抬头→眉头微皱")。3s 以上的镜头至少写 2 个递进动作;2-3s 的短镜只写 1 个动作到位,写多了模型做不完。
  66. - **时间内变化**:光线、衣物飘动、落叶飞舞、表情转换、色温过渡…
  67. - **构图**:主体位置、前后景。
  68. **每镜写完自检(视频化三要素,缺一重写)**:① 至少 1 个强动词;② 至少 1 个镜头运动词;③ 至少 1 个环境/光线动态。
  69. ❌ 静帧式(错):「武松站在山路上,夕阳余晖洒在身后,目光锐利」
  70. ❌ 夹叙事/音效(错):「…夕阳从山脊缓缓下沉;音效:碎石脚步、山风。剧情:醉态登场,为猛虎出场铺垫」
  71. ✅ 纯画面视频式(对):「武松大步从画面深处走来,长袍下摆被风向后吹起,镜头从地面缓慢上摇至全景,夕阳从山脊缓缓下沉」
  72. 镜头之间要有**景别/机位/情绪的变化与剧情递进**,不要每镜雷同——但递进靠画面本身呈现,不靠文字解说;变化靠设计,不靠多切。
  73. ### 第 6 步 · 控制总字数(硬约束 ≤4000 字)
  74. - 全部输出(资产 T2I 提示词 + 多镜头视频提示词,不含表头说明)**总字数 ≤ 4000 字**。
  75. - 预算:每镜约 `3000 ÷ 镜头数` 字,资产部分 ≤1000 字。
  76. - 若超字数:优先压缩重复的环境/风格描述,合并次要镜头(顺便改善节奏),不要靠删运镜或动作过程来省字。
  77. ### 第 7 步 · 交付前质检(逐项核对)
  78. - [ ] 已明确总时长(默认 20-30s,≤30s)与镜头数,各镜时长之和 = 总时长。
  79. - [ ] 每镜时长与其镜头类型匹配(特写/快切 1-3s,中景/叙事 3-5s,全景/固定镜头/长镜头 5-7s)。
  80. - [ ] 镜头数由故事节奏推出,不是拿平均值除出来的。
  81. - [ ] 时长有起伏,不是通篇同一节奏;快剪组后有较长镜头收住;每次切镜都有叙事理由。
  82. - [ ] 已给出所有主体与场景的资产图 T2I 提示词;主体用干净背景只展示主体,场景排除人物;措辞与档案锚点逐字一致。
  83. - [ ] 视频提示词严格按统一模板:`主体:` / `场景:` / `音乐:不要生成任何背景音乐,只生成对应音效` / `镜头N | …`。
  84. - [ ] 头部主体、场景填的是资产短名,镜头正文也只用短名 + 变化点,不逐字复述固定外观。
  85. - [ ] 镜头正文**只有画面**:无音效描述、无 `剧情:…`。
  86. - [ ] 每镜含强动词 + 镜头运动 + 环境/光线动态,无任何一镜读起来像静帧。
  87. - [ ] 多主体已注明关系与出场顺序,每镜标了在场主体;多场景每镜标了场景。
  88. - [ ] 风格关键词全片统一,只在镜头1点明一次。
  89. - [ ] 输出总字数 ≤ 4000 字。
  90. ## 输出格式
  91. 按 `references/templates.md` 输出,含完整范例与风格/运镜/光线词库速查。结构:
  92. 1. **【脚本总览】**:总时长 / 镜头数 / 镜头数理由(节拍如何对应镜头类型)/ 节奏骨架(各镜类型与时长序列)/ 风格 / 主体档案 / 场景锚点。叙事与镜头规划的说明**只写在这里**;
  93. 2. **【资产图文生图提示词】**:主体资产 + 场景资产;
  94. 3. **【视频提示词】**:严格套用统一模板的单条提示词(可直接整块复制),​**纯画面描述**;
  95. 4. 文末标注当前总字数,确保 ≤4000。
  96. 统一模板(不可改动结构与措辞):
  97. ```
  98. 主体:主体1、主体2
  99. 场景:场景名
  100. 音乐:不要生成任何背景音乐,只生成对应音效
  101. 镜头1 | xxxx
  102. 镜头2 | xxxx
  103. ```
  104. ## 硬约束(必须遵守)
  105. 1. 每次必给**主体 + 场景**的资产图文生图提示词(主体干净背景只展示主体;场景展示代表性全貌、排除人物)。
  106. 2. 默认总时长 20-30s,上限 30s;各镜时长之和 = 总时长。
  107. 3. 镜头数完全由故事和节奏决定,无固定档位、无数量上限;每镜时长按镜头类型给(特写/快切 1-3s,中景/叙事 3-5s,全景/建立 5-7s),30s 通常落在 8-14 镜。
  108. 4. 视频提示词严格套用统一模板,音乐行原文不变:`音乐:不要生成任何背景音乐,只生成对应音效`。
  109. 5. 一致性:头部与各镜只用资产短名 + 本镜变化点,不复述固定外观(外观交给资产图)。
  110. 6. 镜头正文**只描述画面**:不写音效、不写剧情作用、不写意图解释。
  111. 7. 每镜是视频提示词不是图片提示词:含强动词、镜头运动、时间内变化三要素。
  112. 8. 输出总字数 ≤ 4000 字。
README.md · README · 282 lines

View this file on GitHub (read at the repository's current version: the scan did not record which commit it read)

  1. <div align="center">
  2. # Seedance 2.5 Prompt.skill
  3. A video prompt Skill built specifically for **Seedance 2.5**.<br>
  4. **Turn an idea / a fragment of a theme / a novel excerpt into a polished 30-second film.**
  5. [![License: MIT](https://img.shields.io/badge/License-MIT-blue.svg)](LICENSE)[![Skill](https://img.shields.io/badge/Format-Skill.md-purple)](./SKILL.md)[![Seedance](https://img.shields.io/badge/Target-Seedance%202.5-success)](https://www.seedance.ai)[![Bilingual](https://img.shields.io/badge/Lang-ZH%20%2F%20EN-orange)](./references/templates.md)
  6. <p align="center">
  7. <a href="./README.md" style="display:inline-block;padding:6px 18px;background:#4f46e5;color:#fff;border-radius:6px 0 0 6px;text-decoration:none;font-weight:600;">English</a> <a href="./README.zh.md" style="display:inline-block;padding:6px 18px;background:#f3f4f6;color:#4b5563;border-radius:0 6px 6px 0;text-decoration:none;font-weight:600;">简体中文</a>
  8. </p>
  9. </div>
  10. <p align="center">
  11. <img src="public/海报.jpg" width="100%" alt="Poster" />
  12. </p>
  13. ---
  14. ## 🌟 Why This Skill?
  15. Writing AI video prompts is a path paved with traps.
  16. It took a long time of stubborn refining before these things became clear:
  17. **1.** Seedance isn't picky about prompts, but it still helps to spell out the **subject, scene, music, and shots** clearly.
  18. - **Subject**: specify exactly which reference image; the name must stay consistent throughout.
  19. - **Scene**: uploading one scene image is enough.
  20. - **Music**: strictly prompt "do not generate any BGM," only sound effects — otherwise post-production editing becomes a nightmare.
  21. - **Shots**: timestamps depend on the content, but the shot numbers must be clearly written, e.g. Shot 01, 02.
  22. **2.** Don't blindly chase resolution. **720P follows prompts best**, 1080P is next, 4K is mediocre.
  23. **3. The core is still picture control.** You need concrete storyboards, framing, transitions, action, camera moves, and plot in your head — good pictures make good stories.
  24. **4.** If you write a prompt you wouldn't even read yourself, don't expect Seedance to produce a high-quality video from it.
  25. These insights hardened into rules, templates, and checklists, and so this Skill was born. **Give it any text — a novel excerpt, a one-line plot, a few theme words — and it returns a multi-shot prompt you can paste straight into Seedance 2.5, along with all the text-to-image prompts for the asset images you need.**
  26. > If you've been tormented by AI video pacing, consistency, and camera work — this Skill is for you.
  27. ---
  28. ## ✨ What It Does
  29. ### 🎬 Direct-to-Seedance 2.5 Prompts
  30. | Input Type | How It's Handled |
  31. |------------|------------------|
  32. | Novel / script / story excerpt | Distills the most visual story beats into consecutive shots |
  33. | A single line of plot | Fills in cause, effect, and turning points, then breaks into shots |
  34. | A few theme words | Creatively expands into a short film with a full arc, then breaks into shots |
  35. Whatever the input, the output is always the same structure: **【Script Overview】+【Asset Image T2I Prompts】+【Video Prompt】**, ready to copy as a single block.
  36. ### 🎯 Default Deliverables
  37. - **Total duration 20–30s** (30s hard cap, auto-collapsed if exceeded)
  38. - **Shot count decided by story rhythm** — no fixed tiers (30s typically lands at 8–14 shots)
  39. - **Every shot carries a time range**, and the sum of all shot durations strictly equals the total
  40. - **All required asset image T2I prompts included**: subjects on clean backgrounds, scenes with people excluded — to lock consistency *before* you generate the video
  41. ### 📐 Skill Workflow
  42. 1. **Parse & expand**: break down story rhythm, inventory subjects & scenes, set the tone
  43. 2. **Set duration & shot count**: lay out shots by story beats, assign duration per shot type, sum and calibrate
  44. 3. **Build subject & scene profiles**: write anchor descriptions, name assets, lock consistency for the whole film
  45. 4. **Write asset image T2I prompts**: one for each subject and scene — generate images first, then video
  46. 5. **Write the multi-shot video prompt**: strictly follow the unified template, describe only moving pictures
  47. 6. **Control word count**: hard constraint ≤ 4000 characters
  48. 7. **Pre-delivery QA**: an 11-point checklist — nothing ships until it passes
  49. ---
  50. ## 🛠️ Design Philosophy
  51. This is the most valuable part of the Skill. Every rule below was earned the hard way.
  52. ### 1. Asset Anchoring: The Real Fix for Consistency
  53. The instinctive way to hold consistency is to "re-describe the protagonist's appearance in every shot" — that's exactly wrong: the more you restate, the more attention the model spends on restating, and the weaker the motion and camera work become.
  54. **The fix**: a subject's and scene's fixed traits are written exactly once in the profile, as the basis for generating asset images (T2I); the video prompt uses only the **short name + this shot's delta** (e.g. "Wu Song": straw hat strapped to his back, hair blown loose by the wind), never re-stating "~30yo brawny man, thick brows…". Consistency is locked by the asset images; the prompt is responsible only for "motion."
  55. ### 2. Configure Shots by Story Type
  56. Many people first decide "I want 12 shots," then divide total duration by 12 to get 2.5s per shot — the result is a flat wall of fast cuts with no rhythm at all. That's chaos cutting.
  57. **Hard rule**: shot count is the *result* of laying out story beats and assigning shot types — never a target you set first. Duration is set by shot type: **close-up / fast cut 2-3s, medium / narrative 3-6s, wide / establishing 6-7s**. Every cut needs a narrative reason (change of space / time / viewpoint / information layer / emotional breakpoint). If one shot can tell it, don't use two. Alternate long and short. After a fast-cut group, settle the rhythm with a longer shot.
  58. Typical 30s skeleton: `Wide establishing (6-7s) → Medium narrative ×2-3 → Close-up fast-cut group (2-3s) ×3-4 → Medium transition → Wide closing`
  59. ### 3. Write Only the Picture, Don't Distract the Model
  60. ```
  61. ❌ Still-frame: Wu Song stands on the mountain path, sunset glow behind him, sharp gaze
  62. ✅ Video: Wu Song strides out from deep in the frame, his long robe's hem blown back by
  63. the wind, the camera slowly tilts up from the ground to a wide shot, the sun slowly
  64. sinks behind the ridge
  65. ```
  66. The shot body describes only "what the camera captures and how it moves" — no sound effects (handled by the fixed Music line in the header), no `Plot: …`, no emotional explanation — without distracting the model.
  67. **The three video-essentials** (self-check after every shot; rewrite if any is missing): ① strong verb ② camera-movement word ③ environment/light dynamic.
  68. ### 4. Unified Prompt Template
  69. ```
  70. Subject: subject1, subject2
  71. Scene: scene name
  72. Music: Do not generate any background music, only generate corresponding sound effects
  73. Shot 1 | 0.0–6.0s | …
  74. Shot 2 | 6.0–9.5s | …
  75. ```
  76. - The Music line is fixed (the single audio instruction for the whole film; no shot describes sound); style keywords are unified film-wide and declared once in Shot 1.
  77. - **Hard constraints**: the sum of all shot durations = total duration ≤ 30s; total output ≤ 4000 characters. When over budget, compress redundant environment/style descriptions and merge minor shots first — **never save words by cutting camera movement**.
  78. ---
  79. ## 📁 Project Structure
  80. ```
  81. .
  82. ├── SKILL.md # Main Skill file (core logic, workflow, hard constraints)
  83. ├── references/
  84. │ └── templates.md # Output template + full example + shot-count decision
  85. │ # cheat sheet + style/camera/light lexicon
  86. ├── public/
  87. │ ├── 关于作者.md
  88. │ ├── 公众号二维码.png
  89. │ └── 海报.jpg
  90. ├── README.md # English (this file)
  91. └── README.zh.md # 简体中文
  92. ```
  93. ---
  94. ## 🚀 Quick Start
  95. ### Who Is This Skill For
  96. Any AI Agent that supports the `Agent Skills` spec (e.g. Codex, Claude Code, WorkBuddy, TRAE Work, Qoder Work, Kimi Work, etc.). As long as your Agent can read and obey a Skill, it can use this to design Seedance 2.5 video prompts.
  97. In principle, models with stronger reasoning are recommended, e.g. Claude Opus 5, K3, DeepSeek-v4-pro, Qwen3.8, GPT-5.6, etc.
  98. ### Installation
  99. ```bash
  100. # Clone the repo
  101. git clone https://github.com/woyin2024/lengyi-seedance2.5-prompt.git
  102. ```
  103. Drop the entire `lengyi-seedance2.5-prompt` folder into your Agent's Skills directory:
  104. - **Claude Code project-level**: `<project>/.claude/skills/lengyi-seedance2.5-prompt/`
  105. - **WorkBuddy user-level**: `~/.workbuddy/skills/lengyi-seedance2.5-prompt/`
  106. - **Other Agents**: follow your tool's Skill installation guide
  107. ### Usage
  108. Once installed, trigger it in conversation any of these ways:
  109. - "Turn this novel excerpt into a Seedance video storyboard"
  110. - "Use the lengyi-seedance2.5-prompt skill to design a 30-second multi-shot prompt, the theme is…"
  111. - "Make a shotlist, image-to-video"
  112. Trigger keywords (Chinese): `seedance 视频提示词`, `多镜头提示词`, `视频分镜脚本`, `镜头脚本`, `shotlist`, `文生视频 / 图生视频`, `AI 视频镜头脚本`.
  113. ### Recommended Workflow
  114. ```
  115. 1. Use the Skill to generate the "Asset Image T2I Prompts" first
  116. 2. Feed the T2I prompts into a text-to-image model (e.g. Image2) to generate
  117. subject / scene asset images
  118. 3. Feed the asset images + the Skill's "Video Prompt" output into Seedance 2.5
  119. ```
  120. ---
  121. ## 🖼️ Output Example
  122. Input — a single line of plot: **"Wu Song, drunk on Jingyang Ridge, encounters a fierce tiger and fights it bare-handed."**
  123. Output (excerpt; full version in [`references/templates.md`](./references/templates.md)):
  124. ```
  125. 【Script Overview】
  126. · Total: 30s · Shots: 9 · Avg: 3.3s
  127. · Shot-count rationale: beats are "entrance — awareness — tiger appears —
  128. dodge — staff breaks — pin down — exhaustion — ending." The opening needs
  129. one 6s wide shot to establish the barren mountain and the drunk at once;
  130. the midsection uses two medium narrative shots; the fight is the emotional
  131. peak, handled with a group of three 2.5s close-up fast cuts; then it
  132. returns to medium-close and wide to settle the rhythm.
  133. · Rhythm skeleton: Wide 6.0 → Medium 3 → Close-up 2 → Medium follow 4.0
  134. → Close-up×3 → Medium-close 3.0 → Wide 3.5
  135. · Style: cinematic wuxia, film grain, realistic texture, warm sunset to cold moon
  136. 【Asset Image T2I Prompts】
  137. - Subject "Wu Song": ~30yo brawny man, thick brows, square face, black beard,
  138. hair tied high, earth-yellow coarse-cloth short martial robe, dark-brown
  139. cloth sash at waist, straw hat on back, holding a wooden staff, full-body
  140. front view, cinematic wuxia realism, film grain, clean white background
  141. showing only the subject, 8K, fine detail, studio soft light
  142. - Subject "Fierce Tiger": adult South China tiger, orange-yellow with black
  143. stripes, sturdy build, eyes glowing green, side-body coiled-strike pose…
  144. - Scene "Jingyang Ridge mountain path": barren mountain dirt road, scattered
  145. rocks and dry grass, dead trees leaning in on both sides, blood-red sunset…
  146. 【Video Prompt】
  147. Subject: Wu Song, Fierce Tiger
  148. Scene: Jingyang Ridge mountain path
  149. Music: Do not generate any background music, only generate corresponding sound effects
  150. Shot 1 | 0.0–6.0s | Jingyang Ridge mountain path, cinematic wuxia, film grain,
  151. realistic texture. Extreme wide shot, low-angle slow push-in, tilting up over
  152. the scattered rocks at the midpoint, settling on a medium shot of the figure
  153. by the end. Subject "Wu Song" stumbles in his steps, flushed with drink,
  154. striding out from deep in the frame, staff dragging on the ground carving a
  155. dirt trail, slowing as he nears and looking around. Long robe hem blown back
  156. by the mountain wind, sun keeps sinking behind the ridge, shadow stretching
  157. ever longer.
  158. Shot 2 | 6.0–9.5s | Medium shot, camera steadily tracking Wu Song's right side.
  159. Wu Song lifts a gourd, tilts his head back to drink, wipes his mouth…
  160. (remaining shots in the full example)
  161. ```
  162. ---
  163. ## 🤝 Contributing
  164. This Skill was born from repeatedly refining something that didn't feel right when I used it — but it can certainly be better.
  165. If you find, while using it, that:
  166. - A certain input type (ads, anime, product showcases…) has room for better shot rhythm
  167. - The lexicon needs additions (new camera, light, or environment-dynamic terms)
  168. - The template could fit Seedance 2.5's latest features more tightly in specific scenarios
  169. - You'd like more output modes (English-only / bilingual)
  170. contributions are very welcome:
  171. 1. Fork this repo
  172. 2. Create a feature branch: `git checkout -b feature/your-feature`
  173. 3. Commit your changes: `git commit -m 'Add some feature'`
  174. 4. Push the branch: `git push origin feature/your-feature`
  175. 5. Open a Pull Request
  176. Extra credit for PRs that include **a real, run-through case** (input → output → model-generated result). That's worth more than any abstract discussion.
  177. ---
  178. ## 📜 License
  179. This project is open-sourced under the [MIT License](LICENSE).
  180. For personal learning use, you're free to use, modify, and distribute it; for commercial use, please obtain authorization in advance — contact via WeChat: lengcp2013.
  181. ---
  182. ## 👤 About the Author
  183. **Leng Yi (冷逸)**, blogger at **沃垠AI (WoyinAI)**, one of China's top AI tech media outlets. A Vibe Coding developer who doesn't write code, obsessed with Prompt, Skills, and Agent engineering.
  184. - **Unified handle across platforms**: 沃垠AI (WoyinAI)
  185. - **Tags**: Product & operations background, in-depth AI reviews, super OPC
  186. - **Content matrix**: WeChat Official Account, Xiaohongshu, Zhihu, GitHub, Bilibili, X, and more
  187. Follow the **沃垠AI** WeChat Official Account for more AI insights, hands-on Prompt engineering, and Skill design methodology:
  188. <p align="center">
  189. <img src="public/公众号二维码.png" width="90%" alt="沃垠AI WeChat Official Account QR Code" />
  190. </p>
  191. ---
  192. <div align="center">
  193. ### If it helped you, leave a ⭐
  194. **Thank you 🙏**
  195. </div>

Ratings and comments

Sign in to rate this skill.

Sign in

Free to use. If it’s helped, pass it on.