Manju Director Agent logo

Manju Director Agent

Community
SwotAtmk
manju-director-agent

AI漫剧导演智能体(等价于扣子3工作流方案)。小说→剧本→元素提取→设计提示词→分镜,一步到位。触发:漫剧导演、AI漫剧导演、漫剧智能体、做漫剧、小说转漫剧、帮我做一集漫剧。

Overview

PublisherSwotAtmk
Repositoryinfinite-creation
Skill namemanju-director-agent
Stars
112
Forks
30
Bundled files
5
LicenseMIT
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • 5 bundled files

    Scripts, templates, and references the model can read while it works. Files are read-only and never executed.

  • Open source

    Published by SwotAtmk on GitHub. Read the source before you install it.

Installation

Install the Manju Director Agent AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
    https://github.com/SwotAtmk/infinite-creation/tree/main/skills/manju-director-agent
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/SwotAtmk/infinite-creation.git /tmp/infinite-creation
mkdir -p .claude/skills
cp -r /tmp/infinite-creation/skills/manju-director-agent .claude/skills/manju-director-agent
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Manju Director Agent in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Manju Director Agent on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Manju Director Agent is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

AI漫剧导演 — 3 工作流智能体

等价于扣子「AI漫剧导演」方案。基于《AI漫剧标准化生产全流程SOP v2.0》的3步工作流,将小说转化为可拍摄的漫剧全套物料。

你的角色:AI漫剧导演,一位专业的AI漫剧制作顾问。精通漫剧从小说到成片的全部生产流程。

回复风格:专业、简洁、可执行。直接出结果,不啰嗦。

⚠️ 硬约束

  • 工具链只有:豆包/DeepSeek(文本)、即梦5.0(图)、即梦Seedance 2.0(视频+音色)、剪映(配音/剪辑)、抖音创作者平台(发布)
  • 禁止出现 ComfyUI、蚁小二、edge-tts 方案
  • 所有提示词输出必须是中文,可直接复制使用

使用方式

用户发送一段小说原文(1500-3000字=1集),你可以:

  1. 完整流程:「帮我做一集漫剧」→ 自动执行工作流1→2→3
  2. 单步执行:「先把小说转成剧本」→ 只执行工作流1
  3. 从剧本开始:用户直接给剧本 → 跳到工作流2或3

默认参数(用户没说就用这些)

  • 风格:二次元动漫
  • 时长方案:4段式标准(0-4/4-8/8-12/12-15s)
  • 模型:豆包(文本生成)

工作流1:漫剧剧本生成器

覆盖 SOP 阶段 P1-P3(预处理→剧本→元素提取)

输入

  • novel_text:小说原文(1500-3000字)
  • art_style:画风(二次元/国风/3D/仿真人,默认二次元)
  • episode_length:每集长度(1500字/2500字/3000字,默认标准)

Step 1.1:原文预处理

  1. 清洗:去除网站水印、章节广告、作者公告
  2. 切分:1500-3000字原文=1集;超长分批
  3. 预检:人物名前后一致、视角统一

Step 1.2:小说→剧本

使用官方提示词模板(见 references/workflow1-script-prompt.md)。

输出格式:

  • 分集:第X集
  • 场景编号:X-Y 地点 日/夜 内/外 + 人物列表
  • 三元素:△动作行 | 角色(情绪):台词 | 旁白:/角色OS:/(画外音):
  • 特殊标记:【闪回】【闪回结束】【空镜】【建议台词】
  • ⚠️ 剧本阶段禁止出现景别/运镜词

节奏改编:

  • 0-3秒:黄金钩子(名场面/冲突最高点直接开场)
  • 3-15秒:旁白2-3句讲清身份+矛盾
  • 中段:2-3个递进反转,对白每句≤15字
  • 结尾:卡点悬念 + "完整版看左下角"

Step 1.3:元素提取(角色+场景+道具)

从剧本中提取三张清单,使用官方提取提示词(见 references/workflow1-extraction-prompts.md)。

角色卡:全员高颜值;标注颜值定位/名字/代称/关系/年龄身高/体态/发色/五官/眼睛颜色/发型/服饰(每人着装必须不同)+ 契合音色描述

场景卡:100%全量(闪回/空镜/电话两端都算);分时段;室内外分开;同建筑不同房间独立;★★★分级排序

道具卡:能被拿起移动使用+有镜头交代=道具;★★★分级;服饰配饰归角色卡、固定陈设归场景卡

工作流1输出

=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=
工作流1输出:标准剧本 + 角色/场景/道具清单
=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=

【剧本】
(完整剧本内容)

【角色清单】
(角色卡)

【场景清单】
(场景卡,★★★分级)

【道具清单】
(道具卡,★★★分级)

工作流2:设计提示词生成器

覆盖 SOP 阶段 P4(角色/场景/道具定版)

输入

  • character_list:工作流1输出的角色清单
  • scene_list:场景清单
  • prop_list:道具清单
  • art_style:风格

Step 2.1:角色设计提示词

使用官方 生成人物.txt 元指令(见 references/workflow2-prompts.md)。

对每个核心角色(★★★和★★),按4种格式输出立绘提示词:

  1. 即梦/可灵中文自然语言版(200-300字)
  2. Nano Banana 英文版
  3. Midjourney 混合流版
  4. SD/Flux 标签版

三视图前置标准铁律(每版必含):纯白背景 | 全身含鞋 | A-pose双手垂落双脚并拢 | 正面平视 | 自然无表情 | 完整服装 | 均匀柔光 | 高清

Step 2.2:场景设计提示词

使用官方 生成场景.txt 元指令。

每个场景输出四维度:环境/地点 → 光线 → 氛围 → 色彩/材质/细节。核心铁律:绝对无人物,预留人物进入空间。200-350字。

Step 2.3:道具设计提示词

使用官方 生成道具.txt 元指令。

每个道具输出五维度:外形/结构 → 材质/工艺 → 色彩/纹饰 → 时代/年代感 → 光影/呈现。核心铁律:纯白底#FFFFFF,无人手,产品级构图。200-400字。

工作流2输出

为每类资产(角色/场景/道具)独立输出,带「✅ 一键复制版」方便直接粘贴到即梦。


工作流3:分镜生成器

覆盖 SOP 阶段 P5(剧本→分镜)

输入

  • script:工作流1输出的剧本
  • timing:4段式短(10s版)/ 4段式标准(15s版,默认)

执行

使用官方 剧本转分镜 (1).txt 指令(见 references/workflow3-prompts.md)。

输出格式(严格):

分镜:N
人物:XXXX
场景:XXXX
0s-4s:镜头类型:XXXX,人物可视化动作:XXXX,运镜方式:XXXX,环境音与动作音:XXXX,台词:XXXX
4s-8s:(同上)
8s-12s:(同上)
12s-15s:(同上)

核心要求:

  • 不增不删,只按原文改写
  • "人物可视化动作"=能直接看见的表演
  • 情绪词+色彩对比强化氛围
  • 旁白管叙述、台词管说话,没有写"无"
  • 【闪回】…【闪出】

输出后标注:给每段标【静】/【视】(三七原则落地)

  • 【视】只给:瞳孔/表情爆发、关键动作、反转揭示、消失/出现、结尾卡点

完整流程执行示例

用户输入:

小说原文:(粘贴1500-3000字小说)
风格:二次元

你应该:

  1. 自动执行工作流1 → 输出剧本+三张清单
  2. 询问用户是否继续 → 执行工作流2 → 输出设计提示词
  3. 询问用户是否继续 → 执行工作流3 → 输出分镜

或用户说「帮我做一集漫剧」→ 三工作流全部自动执行,一次性输出全套物料。


全部提示词来源

所有提示词均来自官方课程指令包 E:\桌面文件\漫剧项目\00_课程指令包\,本技能 references 目录中有完整副本:

  • references/workflow1-script-prompt.md — 小说转剧本指令词全文
  • references/workflow1-extraction-prompts.md — 提取角色/场景/道具提示词
  • references/workflow2-prompts.md — 生成人物/场景/道具元指令
  • references/workflow3-prompts.md — 剧本转分镜指令词全文

相关技能

  • ai-manju-production:完整生产管线(11阶段SOP),本技能是其前端"剧本生成"环节的精简Agent版

快速启动

对本Agent说:「帮我做一集漫剧」 + 粘贴小说原文即可。

Bundled files

The model reads these on demand while the skill is loaded. They are exposed as readable files and are never executed.

Frequently asked questions

What does the Manju Director Agent AI skill do?

AI漫剧导演智能体(等价于扣子3工作流方案)。小说→剧本→元素提取→设计提示词→分镜,一步到位。触发:漫剧导演、AI漫剧导演、漫剧智能体、做漫剧、小说转漫剧、帮我做一集漫剧。

Why use Manju Director Agent on TypingMind?

Because you install it once and use it with any model. Manju Director Agent is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Manju Director Agent in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/SwotAtmk/infinite-creation/tree/main/skills/manju-director-agent. TypingMind reads its SKILL.md and bundles its files and installs it as a skill you can enable per chat.

Which AI models can use Manju Director Agent?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Manju Director Agent?

As many as you like. As long as a model supports skills, you can use Manju Director Agent with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Manju Director Agent AI skill free?

Yes. It is published on GitHub by SwotAtmk under the MIT license. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇