Skip to content
Back to search

ai-visual-director

SkillWorks with Claude CodeGitHub

"AI视觉导演主控路由。用于 /create /avd/create 故事到视频、角色卡、场景图、故事板、分镜图、视频 Prompt、海报、Prompt QC、状态锁定和子技能分发。普通故事默认 /create standard。"

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/jijiutong--ai-visual-director/badge.svg)](/listings/jijiutong--ai-visual-director)

HTML

<a href="/listings/jijiutong--ai-visual-director"><img src="/api/listings/jijiutong--ai-visual-director/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.

These steps run a script from the network that findsafeskills does not scan.

npx ai-visual-director

You can also use the install script or install from source:

curl -fsSL https://raw.githubusercontent.com/jijiutong/ai-visual-director/main/install.sh | sh
git clone https://github.com/jijiutong/ai-visual-director.git
cd ai-visual-director
sh install.sh

Restart Claude Code, then enter:

/create Two swordsmen face off before a Buddha statue in a rain-soaked temple. The master realizes his disciple has fallen to darkness, and both draw their swords. 15s, 7 shots.

The default is `/create standard`:

Story intake → Shot budget → Character anchor → Scene anchor
→ Storyboard → Video prompt → QC → Execution checklist

See the [User Guide](./docs/usage.md) for the complete workflow.

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 · 135 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: ai-visual-director
  3. description: "AI视觉导演主控路由。用于 /create /avd/create 故事到视频、角色卡、场景图、故事板、分镜图、视频 Prompt、海报、Prompt QC、状态锁定和子技能分发。普通故事默认 /create standard。"
  4. ---
  5. # AI Visual Director — 主控路由
  6. 这是给 agent 执行用的主 Skill,不是产品 README。README 负责介绍、安装和能力展示;本文件只定义触发、路由、必读文件和执行约束。
  7. ## 触发条件
  8. 当用户提出以下任一需求时,使用本 Skill:
  9. - 把故事、剧本、小说片段、分镜想法转成可执行视觉方案
  10. - 生成角色卡、场景图、故事板、分镜图、视频 Prompt、海报或执行包
  11. - 使用 `/create` `/source` `/storyboard` `/character` `/scene` `/video` `/style` `/poster`
  12. - 使用 `/avd/*` 命名空间命令
  13. - 要求进行角色一致性、场景一致性、参考图锚定、视频平台适配或 Prompt QC
  14. ## 默认行为
  15. 如果用户只粘贴故事或说“一键生成/直接出/auto”,默认路由到:
  16. ```text
  17. /create standard
  18. ```
  19. 默认输出状态为 `draft`。只有用户明确确认、锁定或提交时,才写回项目状态:
  20. ```text
  21. draft -> locked -> committed
  22. ```
  23. ## 命令路由
  24. | 用户意图 | 路由 | 必读文件 |
  25. |---|---|---|
  26. | 一键故事到视频包 | `/create` 或 `/avd/create` | `sub-skills/create/SKILL.md` |
  27. | 读取 Obsidian/Markdown/粘贴/批处理 | `/source` 或 `/avd/source` | `sub-skills/source/SKILL.md` |
  28. | 角色外观、表情、服装、DNA、一致性 | `/character` 或 `/avd/character` | `sub-skills/character/SKILL.md` |
  29. | 场景空间、光照、材质、地理关系 | `/scene` 或 `/avd/scene` | `sub-skills/scene/SKILL.md` |
  30. | 故事板、分镜图、镜头设计、全案板 | `/storyboard` 或 `/avd/storyboard` | `sub-skills/storyboard/SKILL.md` |
  31. | 视频 Prompt、首尾帧、平台参数、拆段 | `/video` 或 `/avd/video` | `sub-skills/video/SKILL.md` |
  32. | 风格浏览、融合、迁移、导演方法参考 | `/style` 或 `/avd/style` | `sub-skills/style/SKILL.md` |
  33. | 独立海报、封面、营销视觉 | `/poster` 或 `/avd/poster` | `sub-skills/poster/SKILL.md` |
  34. | 台词、字幕、口型节奏 | `/dialogue` | `engines/dialogue-engine.md` |
  35. | 环境音、拟音、音乐、混响 | `/sound` | `engines/sound-engine.md` |
  36. | 锁定、提交、解锁、检查状态 | `/lock` `/commit` `/unlock` `/check` | `engines/state-commit.md` |
  37. 先读路由目标文件,再按该文件继续读取模板、规则和引擎。不要只凭本文件生成最终资产。
  38. `/create` 不替代 `/storyboard`、`/character`、`/scene`;它只在一键总编排中按模式调度三条核心资产子链。
  39. ```text
  40. /storyboard 核心资产:故事板
  41. /character 核心资产:角色卡
  42. /scene 核心资产:场景图
  43. ```
  44. ## 主执行链
  45. `/create` 是默认总编排链,按需调度角色、场景、分镜和视频模块。具体步骤以 `sub-skills/create/SKILL.md` 的执行链路为权威,本文件只列主控边界:
  46. ```text
  47. engines/command-gate.md
  48. -> sub-skills/create/SKILL.md
  49. -> rules/format-contract.md
  50. -> rules/prompt-qc.md
  51. -> rules/final-video-qc.md
  52. -> engines/state-commit.md
  53. ```
  54. ## 必须遵守
  55. - 输出资产前必须读取对应子 Skill 和模板文件。
  56. - 格式边界以 `rules/format-contract.md` 为准。
  57. - 编号体系以 `rules/numbering.md` 为准。
  58. - 视频负面词和质量检查以 `rules/qc.md`、`rules/final-video-qc.md`、`rules/prompt-qc.md` 为准。
  59. - 角色一致性以 `rules/character-consistency.md` 为准。
  60. - 场景一致性以 `rules/scene-consistency.md` 为准。
  61. - 默认参数以 `rules/one-click-defaults.md` 和 `api-config.template.env` 为准。
  62. - 用户没有确认时,不写回 `state/` 主状态;允许在输出中生成 draft 状态块或执行清单。
  63. ## 模板读取规则
  64. | 输出类型 | 必读模板 |
  65. |---|---|
  66. | 角色卡 | `templates/character-sheet.md` |
  67. | 场景图 | `templates/scene-card.md` |
  68. | 全案板 | `templates/full-board.md` |
  69. | 分镜图 | `templates/quick-board.md` |
  70. | 海报 | `templates/poster.md` |
  71. | 台词脚本 | `templates/dialogue-script.md` |
  72. | 声音设计 | `templates/sound-design-sheet.md` |
  73. 模板是输出结构的权威源。模板内已有的模块数、参数数、版式编号和禁止项,不要在本文件重复维护。
  74. ## 能力分层
  75. | 层级 | 内容 | 默认策略 |
  76. |---|---|---|
  77. | A 稳定治理层 | command-gate、format-contract、lock-state、prompt-qc、auto-repair | 所有命令必走 |
  78. | B 资产生产层 | character、scene、storyboard、video、dialogue、sound、poster | 按用户意图调用 |
  79. | C 导演增强层 | shot-budget、video-director、motion-physics、emotion-curve、color-narrative | project/standard/full 默认可用 |
  80. | D 探索发散层 | fusion、multi-version、style-migration、director-imitation、series、mood-slider | 用户明确要求才启用 |
  81. ## 写回策略
  82. | 状态 | 触发 | 行为 |
  83. |---|---|---|
  84. | `draft` | 默认生成 | 只输出草稿、状态块或执行清单,不改 `state/` 主状态文件 |
  85. | `locked` | 用户确认锁定 | 写入 `state/lock-state.md`,禁止自动覆盖 |
  86. | `committed` | 用户确认提交 | 写回 `state/variable-registry.md`、`state/asset-map.md`、`state/project-graph.md` 等持久状态 |
  87. 涉及状态变更时,先读 `engines/state-commit.md` 和相关 `state/*.md`。
  88. ## 禁止事项
  89. - 禁止把 README 的安装说明、营销文案、能力展示复制进本文件。
  90. - 禁止不读子 Skill 或模板,直接凭记忆输出角色卡、场景图、故事板或视频 Prompt。
  91. - 禁止把全案板和分镜图混为同一种输出;先用 `rules/format-contract.md` 判断 `output_type`。
  92. - 禁止默认开启 D 类探索能力,除非用户明确要求发散、多版本、融合、迁移或导演模仿。
  93. - 禁止在用户未确认时覆盖已 locked/committed 的状态。
  94. - 禁止输出未经过 Prompt QC、格式合同和视频 QC 的最终执行包。
  95. ## 全局负面清单
  96. 所有图像/视频 Prompt 必须按需合并 `rules/negative-prompt.md`,并至少包含:
  97. ```text
  98. no watermark, no logo, no random large text, no garbled Chinese, no broken faces,
  99. no duplicated limbs, no messy panels, no low-quality collage, no text overlay,
  100. no speech bubbles (unless manga format), no cartoon style (unless specified),
  101. no flat illustration, no marketing poster style
  102. ```
README.md · README · 220 lines · 1 finding marked

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

  1. # AI Visual Director
  2. <p align="center">
  3. <strong>Turn Stories into Production-Ready Visual Plans</strong>
  4. </p>
  5. <p align="center">
  6. From character and scene anchoring to storyboard design, video prompts, platform adaptation, and quality control.<br>
  7. A Prompt Production OS for AI films, short dramas, motion comics, and serialized content.
  8. </p>
  9. <p align="center">
  10. <a href="./README.md"><strong>English</strong></a>
  11. ·
  12. <a href="./README.zh-CN.md">中文</a>
  13. ·
  14. <a href="./docs/usage.md">User Guide</a>
  15. ·
  16. <a href="./docs/capabilities.md">Capability White Paper</a>
  17. ·
  18. <a href="./docs/system-architecture.md">System Architecture</a>
  19. </p>
  20. <p align="center">
  21. <img alt="License: MIT" src="https://img.shields.io/badge/License-MIT-111827">
  22. <img alt="Claude Code Skill" src="https://img.shields.io/badge/Claude_Code-Skill-D97757">
  23. <img alt="Prompt Production OS" src="https://img.shields.io/badge/Prompt-Production_OS-0F766E">
  24. </p>
  25. ---
  26. > A story goes in; character sheets, scene references, storyboards, video prompts, and an execution package come out.
  27. > This is not an inspiration toy. It is a visual production line with command gates, format contracts, state locking, asset separation, and multidimensional QC.
  28. ```text
  29. Story
  30. → Intake
  31. → Director Decisions
  32. → Character & Scene Anchors
  33. → Storyboard
  34. → Platform Prompts
  35. → Quality Control
  36. → Execution Package
  37. ```
  38. ## Why It Exists
  39. AI image and video models are already powerful, but production is about more than generating a single beautiful image:
  40. - A character's face, costume, props, and posture drift between shots
  41. - Scene geometry, lighting direction, and materials lose continuity
  42. - Presentation boards, marketing art, and video references get mixed together and introduce noise
  43. - Prompts grow longer after repeated revisions while creative boundaries become less clear
  44. - Reference-image, first/last-frame, and text rules cannot be reused directly across video platforms
  45. AI Visual Director encodes the director's workflow as an inspectable constraint system:
  46. ```text
  47. Full capability preserved
  48. → Default authority constrained
  49. → Exploration explicitly triggered
  50. → State committed only after confirmation
  51. → Video assembled, never redesigned
  52. ```
  53. ## Core Capabilities
  54. | Capability | Coverage |
  55. |---|---|
  56. | **53 visual styles** | Film, animation, Eastern aesthetics, science fiction, glitch art, and director-method references |
  57. | **44 layout systems** | Full boards, character boards, scene boards, timelines, HUDs, vertical formats, and keyframe sequences |
  58. | **140+ camera techniques** | Shot sizes, camera movements, angles, focal lengths, aspect ratios, transitions, and special POVs |
  59. | **19 numbering systems** | Automatic mapping for style, emotion, color, expression, body language, environment, material, sound, and more |
  60. | **8 character-consistency methods** | Character sheets, five-angle views, expression sets, costume and weapon details, DNA, and reference anchoring |
  61. | **7 scene-consistency methods** | All-in-one boards, nine-grids, panoramas, blueprints, surround captures, and widescreen anchors |
  62. | **5 video platforms** | Platform-specific prompts for Seedance, Runway, Kling, Luma, and Pika |
  63. | **12 video QC checks** | Character, scene, lighting, props, action, transitions, subtitles, sound, and asset purpose |
  64. ## Start in 30 Seconds
  65. ### Install
  66. ```bash
  67. npx ai-visual-director
  68. ```
  69. You can also use the install script or install from source:
  70. ```bash
  71. curl -fsSL https://raw.githubusercontent.com/jijiutong/ai-visual-director/main/install.sh | sh
    • Note: Pipes a downloaded script directly into a shell (curl | sh / wget | bash)
    • Line 90 on GitHub
  72. ```
  73. ```bash
  74. git clone https://github.com/jijiutong/ai-visual-director.git
  75. cd ai-visual-director
  76. sh install.sh
  77. ```
  78. Restart Claude Code, then enter:
  79. ```text
  80. /create Two swordsmen face off before a Buddha statue in a rain-soaked temple. The master realizes his disciple has fallen to darkness, and both draw their swords. 15s, 7 shots.
  81. ```
  82. The default is `/create standard`:
  83. ```text
  84. Story intake → Shot budget → Character anchor → Scene anchor
  85. → Storyboard → Video prompt → QC → Execution checklist
  86. ```
  87. See the [User Guide](./docs/usage.md) for the complete workflow.
  88. ## Choose Your Entry Point
  89. | What you want to do | Command | Learn more |
  90. |---|---|---|
  91. | Generate a complete execution package from a story | `/create` | [One-click orchestration](./sub-skills/create/SKILL.md) |
  92. | Read from Obsidian, Markdown, or pasted content | `/source` | [Input source system](./sources/README.md) |
  93. | Lock character appearance and character DNA | `/character` | [Character workflow](./sub-skills/character/SKILL.md) |
  94. | Lock scene geometry, materials, and lighting | `/scene` | [Scene workflow](./sub-skills/scene/SKILL.md) |
  95. | Design shots, storyboards, and full boards | `/storyboard` | [Storyboard workflow](./sub-skills/storyboard/SKILL.md) |
  96. | Assemble multi-platform video prompts | `/video` | [Video workflow](./sub-skills/video/SKILL.md) |
  97. | Design dialogue, subtitles, and lip-sync rhythm | `/dialogue` | [Dialogue engine](./engines/dialogue-engine.md) |
  98. | Design ambience, foley, music, and reverb | `/sound` | [Sound engine](./engines/sound-engine.md) |
  99. | Explore styles, fusion, or director methods | `/style` | [Style workflow](./sub-skills/style/SKILL.md) |
  100. | Generate a standalone marketing poster | `/poster` | [Poster workflow](./sub-skills/poster/SKILL.md) |
  101. Governance commands: `/lock` · `/commit` · `/unlock` · `/check` · `/repair`
  102. ## Three Production Tiers
  103. | Capability | `fast` | `standard` | `full` |
  104. |---|:---:|:---:|:---:|
  105. | Story intake and shot budget | ✓ | ✓ | ✓ |
  106. | Character and scene anchoring | Inline DNA | Standard templates | Standard templates |
  107. | Storyboard and video prompts | Condensed | Complete | Complete |
  108. | Dialogue and sound design | - | On demand | ✓ |
  109. | Full board and complete QC | - | ✓ | ✓ |
  110. | Best for | Rapid validation | Production | Full delivery |
  111. ## How the System Stays Stable
  112. The system uses a four-layer capability model. Creative capabilities remain available, but each layer has different default authority:
  113. | Layer | Invocation policy | Responsibility |
  114. |---|---|---|
  115. | **Stable governance layer** | Always on | Routing, format contracts, lock state, prompt QC, and automatic repair |
  116. | **Asset production layer** | Command-driven | Characters, scenes, storyboards, video, dialogue, sound, and posters |
  117. | **Director enhancement layer** | Project mode | Shot budget, pacing, emotion curves, color narrative, and motion physics |
  118. | **Exploratory layer** | Explicit trigger | Multiple versions, style fusion, migration, director methods, and series continuation |
  119. State lifecycle:
  120. ```text
  121. draft Default draft; does not pollute primary state
  122. ↓
  123. locked Confirmed by the user; cannot be overwritten automatically
  124. ↓
  125. committed Persisted to formal project state
  126. ```
  127. Learn more: [Capability Boundaries](./docs/capabilities.md) · [State Management](./state/README.md) · [Format Contract](./rules/format-contract.md)
  128. ## Architecture Overview
  129. ```mermaid
  130. flowchart LR
  131. A[User Story] --> B[Source & Story Intake]
  132. B --> C[Command Gate & Format Contract]
  133. C --> D[Director & Asset Planning Engines]
  134. D --> E[Character / Scene / Storyboard Assets]
  135. E --> F[State & Consistency Anchors]
  136. F --> G[Platform Prompt Assembly]
  137. G --> H[Scoring / QC / Auto Repair]
  138. H --> I[Execution Package]
  139. ```
  140. | Module | Role | Documentation |
  141. |---|---|---|
  142. | `SKILL.md` | Main routing, command boundaries, and execution contract | [Main skill](./SKILL.md) |
  143. | `sources/` | Paste, Markdown, Obsidian, and frontmatter intake | [Input sources](./sources/README.md) |
  144. | `engines/` | Routing, director decisions, planning, scoring, repair, and packaging | [Engine map](./engines/README.md) |
  145. | `docs/` | User guide, capability white paper, architecture, and command reference | [Documentation center](./docs/README.md) |
  146. | `state/` | State registry, asset mappings, shot snapshots, and project dependency graphs | [State management](./state/README.md) |
  147. | `rules/` | Format, consistency, numbering, quality, and platform constraints | [Quality rules](./rules/qc.md) |
  148. | `templates/` | Character, scene, storyboard, poster, and sound templates | [Template directory](./templates/README.md) |
  149. | `knowledge/` | Camera, lighting, composition, material, performance, and sound knowledge bases | [Knowledge base](./knowledge/README.md) |
  150. | `platforms/` | Image and video platform adaptation and API integration | [Platform adapters](./platforms/README.md) |
  151. | `sub-skills/` | Independent user-facing command workflows | [Sub-skills](./sub-skills/README.md) |
  152. | `imitation/` | Director and studio visual-method reference library | [Style references](./imitation/README.md) |
  153. | `examples/` | Sample projects and reference outputs | [Examples](./examples/) |
  154. | `projects/` | Persistence for films, series, and multi-segment projects | [Project system](./projects/README.md) |
  155. | `.agents/` | Agent configuration directory | — |
  156. | `.claude/` | Claude Code configuration directory | — |
  157. For the complete data flow and file-placement rules, see [System Architecture](./docs/system-architecture.md).
  158. ## Documentation Center
  159. | Document | Best for | What you get |
  160. |---|---|---|
  161. | [User Guide](./docs/usage.md) | First-time users and command lookup | Command selection, production tiers, common workflows, and state commits |
  162. | [Capability White Paper](./docs/capabilities.md) | Understanding system boundaries | Four-layer model, asset purposes, defaults, and invocation conditions |
  163. | [System Architecture](./docs/system-architecture.md) | Maintainers and contributors | Execution flow, module roles, directory structure, and extension principles |
  164. | [Main Skill Contract](./SKILL.md) | Skill developers | Main routing, governance rules, module mapping, and global negatives |
  165. | [Documentation Index](./docs/README.md) | Everyone | Quick navigation from tasks to authoritative documentation |
  166. ## Platform Coverage
  167. | Image generation | Video generation |
  168. |---|---|
  169. | GPT Image · Midjourney · DALL-E · SDXL / SD3 | Seedance · Runway · Kling |
  170. | Flux · Ideogram · Tongyi Wanxiang · Recraft | Luma · Pika · ComfyUI / IP-Adapter extensions |
  171. See [Platform Adaptation](./platforms/README.md) for configuration and limitations.
  172. ## License
  173. [MIT](./LICENSE)

Ratings and comments

Sign in to rate this skill.

Sign in

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