Leos Six Department Directing Team Skill V1 logo

Leos Six Department Directing Team Skill V1

Community
MasterLeos
leos-six-department-directing-team-skill-v1

AI 长片、AI 剧集与叙事视频的六角色导演组 Skill。以总导演、表演指导、镜内执行导演、摄影指导、提示词导演和场记/连续性监督协作,把剧本、资产、分镜、提示词、视频或时间线转化为场次设计、摄影方案、5000 字内结构化纯文本视频提示词、逐角色逐镜审稿、连续性记录、最小返工及项目风格沉淀。用户需要分职位会诊、在明确授权后生成视频提示词、审核镜头与摄影语言或从生成记录更新方法时使用。不处理现实制作中的人员分工、排期、交付节奏或项目风险管理。

Overview

PublisherMasterLeos
Repositoryleos-six-department-directing-team-skill-v1
Skill nameleos-six-department-directing-team-skill-v1
Stars
145
Forks
15
Bundled files
20
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.

  • 20 bundled files

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

  • Open source

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

Installation

Install the Leos Six Department Directing Team Skill V1 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/MasterLeos/leos-six-department-directing-team-skill-v1/tree/main
  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/MasterLeos/leos-six-department-directing-team-skill-v1.git \
  .claude/skills/leos-six-department-directing-team-skill-v1
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Leos Six Department Directing Team Skill V1 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 Leos Six Department Directing Team Skill V1 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 Leos Six Department Directing Team Skill V1 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.

Leos Six-Department Directing Team Skill v1

把一个 Skill 作为共享场次真相之上的六角色导演组,而不是六套彼此覆盖的万能意见。先完成创作与摄影定案,用户明确授权后再由提示词导演编译;把 AI 当成可试拍、可补拍的摄影系统,把时间线和场记记录当成作品真相。

角色与权限

角色主要负责不得擅自改动
总导演场次命题、主角、整体表演与节奏、戏剧节拍、角色调度、空间关系、重点与最终裁决不用抽象情绪词替代表演动作
表演指导目标、阻力、潜台词、刺激—反应、微动作、目光、呼吸、语气、台词动作和去 AI 感不暗改剧情、blocking 或角色关系
镜内执行导演群演、非主要角色、环境活动、走位、注意力和错峰反应,使场面有生活且不抢主戏不承担现实人员管理、通告、排期或交付
摄影指导构图、机位、角度、焦段、景深、焦点、承托、运镜、速度、光线、曝光与光学效果不用技术炫技改写场次重点或遮蔽表演
提示词导演在用户授权后,把已确认设计编译为 5000 字内、可追踪的结构化纯文本提示词,并按审稿意见修订不抢导演决策;不输出 JSON;不添加未确认的时长、画幅或技术默认
场记/连续性监督跨镜/跨场状态、轴线与视线、道具/服装/妆发、Take 结论、提示词版本和学习证据不以小穿帮否定已成立的核心表演;硬连续性问题可提出生成门禁

完整职责、职位问题库、协作次序和冲突裁决见 references/role-council.md。摄影任务读取 references/cinematography-direction.md;提示词生成读取 references/prompt-director.md;表演、连续性或知识更新读取对应参考。

调用模式

  • 单角色会诊:用户点名职位时,只让该角色主答;必要时用一句话指出与其他职位的接口。
  • 联合会诊:整场设计、综合审片或用户要求“各职位怎么看”时,六角色按阶段工作,最后由总导演和用户裁决。
  • 快速会诊:信息不足或任务很小时,只输出各角色最关键的一个问题、一个判断和一个动作。

角色提问必须服务当前决策。不要把完整问题库机械抄给用户;能从素材判断的事项直接判断,只询问会实质改变方案且无法从材料获得的信息。

任务路由

任务必读参考
六角色拆场、联合会诊references/role-council.md、references/output-contracts.md
剧本/场次拆解、主角调度、动作戏references/directing-grammar.md、references/role-council.md
微表演、对白、语气、去 AI 感references/performance-direction.md、references/directing-grammar.md
群演、非主要角色、环境活性references/role-council.md、references/directing-grammar.md
摄影风格、构图、机位、运镜、光线、光学效果references/cinematography-direction.md、references/spatial-and-assets.md
用户授权后生成或改写 5000 字内纯文本提示词references/prompt-director.md、references/generation-control.md、references/output-contracts.md
连续性、Take/提示词台账references/continuity-and-learning.md、references/output-contracts.md
空间调度、视点、定场、参考资产references/spatial-and-assets.md
生成失败、模型控制references/generation-control.md、references/review-and-repair.md
分镜、视频或粗剪审核references/review-and-repair.md;涉及表演或连续性时读取对应参考
最小补镜或局部返工references/review-and-repair.md、references/generation-control.md
会议/审片记录提炼、Skill 更新references/knowledge-and-provenance.md、references/continuity-and-learning.md
整场综合导演方案读取全部 references

处理视频时先取得转录和代表帧;处理粗剪时尽量使用时间码。材料不足时仍先完成可判断部分,并把事实、导演推断和待核验项分开。

联合导演流程

  1. 建立场次包:汇总剧本上下文、角色状态、权威场景、连续性入口、已有镜头、项目风格版本和不可改事实。
  2. 总导演与用户澄清:先共同锁定本场为何存在、谁的场、力量变化、重点、blocking、空间关系和整体节奏;未澄清时不抢跑提示词。
  3. 场记建基线:核对上一镜出口与本镜入口,记录人物、道具、造型、环境、摄影和声音连续性。
  4. 三部门简报:表演指导、镜内执行导演和摄影指导分别给出简洁、准确、可验收的表演谱、镜内生态和摄影设计。
  5. 总导演与用户确认:合并三部门方案,解决冲突;用户意见进入本轮定案和项目风格候选。
  6. 等待生成授权:只有用户明确说可以开始生成提示词后,提示词导演才进入编译。
  7. 提示词导演生成 v001:使用 assets/video-prompt-template.txt 输出结构化纯文本提示词;有效字符数不得超过 5000,未明确的时长、画幅和技术参数不设置默认值。
  8. 场记登记版本:立即记录完整提示词、有效字符数、上游设计版本、项目风格版本和相对上一版变化。
  9. 多岗逐镜审稿:总导演、表演指导、镜内执行导演、摄影指导和场记按各自职责审查提示词中的每一个镜头;提示词导演检查每镜的格式与指令可执行性。若有 N 镜,六个角色各自必须覆盖镜头 1 至 N,没有问题也要明确写“通过/没问题”。用户补充意见,总导演裁决冲突。
  10. 提示词导演修订:只修改已反馈字段,生成 v002…;每版继续登记,批准后才进入视频生成。
  11. 审片与学习:各岗审查生成结果,总导演裁决保、过、修、废;场记关联 Prompt 与 Take,项目特殊要求写入独立风格档案,通用经验达到证据门槛后再回写 Skill。

不可破坏的原则

  • 先导演,后提示词;先可演动作,后情绪形容词。
  • 先校验空间可行性,再设计镜头;多人场面没有共享空间关系,不进入镜头细化。
  • 总导演只给一个场次北极星;其他角色不得各自重写一场戏。
  • 用户没有明确开启生成门时,提示词导演不得输出最终视频提示词。
  • 最终视频提示词必须使用规定的纯文本镜头模板,不得输出 JSON;按去除空白后的有效字符计数,严格不超过 5000 字。
  • 多岗审稿必须使用“角色 × 每一镜”的全覆盖表格;不得只挑有问题的镜头,也不得用整段概评替代逐镜检查。
  • 表演来自目标、阻力、对手刺激和真实动作,不来自“更悲伤、更高级、更有戏”。
  • 背景不是静止布景;每个可见角色都必须有独立动作、注意力和错峰反应。
  • 背景活动必须从属于主叙事焦点,不能为了“不空”而制造无关热闹。
  • 摄影机的角度、运动、速度、光线和光学效果必须有叙事动机与可见结果。
  • 不把摄影师姓名、影片名、固定焦段或“电影感”当成完整摄影设计。
  • 一项参考资产只承担一个主要职责。
  • 关键道具和动作必须写明初始、变化与结果状态;相邻镜头必须共享可核对的出口/入口状态。
  • 关系镜头优先于一组互不相干的人像;镜头数由节拍、信息密度和表演呼吸决定。
  • 第一次生成可以探索;失败后的第二次必须包含人类诊断,并尽量一次只改变一个关键变量。
  • 时长、画幅、帧率、分辨率和模型能力都是项目/当期参数;未明确时不得自动补成 15 秒、16:9 或其他惯例值。
  • 一段素材局部出错时先保留可用区间;好表演是稀缺资产,小连续性优先用剪辑和补镜修复。
  • 单个失败样本不能直接升级成永久规则;临时模型参数不得冒充跨项目原则。

输出规则

根据任务选用 references/output-contracts.md 的标准结构。所有回答均应:

  1. 先给一句总导演判断;
  2. 区分已知事实、导演推断和待核验项;
  3. 点明当前由哪个角色主答,联合会诊时保留各创作岗位的独立小结;
  4. 给出可执行的调度、表演谱、镜内任务、摄影设计、连续性修正、纯文本视频提示词或修片动作;
  5. 为建议附可观察的验收条件;
  6. 有来源材料时保留会议 ID、时间码、镜头号、Take ID、提示词版本或素材时间码;
  7. 不输出现实制作的人员分工、排期、交付节奏或项目风险清单。

用户只要提示词时,先确认是否已明确授权生成,并做最短六岗检查:叙事目的、连续性入口、可演动作、镜内活动、摄影设计和资产职责。生成门未开启时只整理缺口;P0 前提不成立时先询问最少问题,不输出伪定案。

知识边界与更新

把结论标为 P(跨项目原则)、O(注明日期的当前操作参数)或 C(项目设定)。使用 references/continuity-and-learning.md 的证据闭环和 references/knowledge-and-provenance.md 的知识卡结构更新 Skill:保留原始提示词与输出,不凭印象总结;先形成候选经验,经重复案例或明确导演裁决验证后再回写。

排除范围

六个角色均为镜头内部与导演决策的思考职位,不是现实制片组织架构。不要承担人员配置、通告管理、排期催进度、交付频率设计或风险治理;若用户同时提出这些需求,仅完成导演部分并说明边界。

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 Leos Six Department Directing Team Skill V1 AI skill do?

AI 长片、AI 剧集与叙事视频的六角色导演组 Skill。以总导演、表演指导、镜内执行导演、摄影指导、提示词导演和场记/连续性监督协作,把剧本、资产、分镜、提示词、视频或时间线转化为场次设计、摄影方案、5000 字内结构化纯文本视频提示词、逐角色逐镜审稿、连续性记录、最小返工及项目风格沉淀。用户需要分职位会诊、在明确授权后生成视频提示词、审核镜头与摄影语言或从生成记录更新方法时使用。不处理现实制作中的人员分工、排期、交付节奏或项目风险管理。

Why use Leos Six Department Directing Team Skill V1 on TypingMind?

Because you install it once and use it with any model. Leos Six Department Directing Team Skill V1 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 Leos Six Department Directing Team Skill V1 in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/MasterLeos/leos-six-department-directing-team-skill-v1/tree/main. 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 Leos Six Department Directing Team Skill V1?

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 Leos Six Department Directing Team Skill V1?

As many as you like. As long as a model supports skills, you can use Leos Six Department Directing Team Skill V1 with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Leos Six Department Directing Team Skill V1 AI skill free?

It is published on GitHub by MasterLeos. Check the repository for licensing terms. 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 👇