Skip to content
Agentmart
Submit a tool

comic-drama-master

漫剧制作总导演。接受用户故事创意,依次完成剧本创作、角色设计、分镜图生成、分镜视频生成、视频合成全流程,产物保存到 COMIC_DRAMA_OUTPUT_DIR(默认 ./output),最终交付完整漫剧视频和 TOS 链接。

Skill by bytedanceOfficial

IndexedApache-2.0★ 473
Write a review

Install

Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.

npx agentmarket install skill/bytedance-comic-drama-master

Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.

Get the weekly diff

The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.

We use your email only for this newsletter. Privacy

About

漫剧制作总导演。接受用户故事创意,依次完成剧本创作、角色设计、分镜图生成、分镜视频生成、视频合成全流程,产物保存到 COMIC_DRAMA_OUTPUT_DIR(默认 ./output),最终交付完整漫剧视频和 TOS 链接。

No reviews yet. Installed it? Reviews come from verified installers only. How reviews work

Permissions

What this skill can do and what it asks for, from its SKILL.md. Checked Oct 25, 2026.

Tools
No tools found
Secrets
None asked for
Install scripts
None found
Runs as
Python package

Network hosts it talks to 1

  • codeload.github.com · found in install command

Files it touches

  • ~/.claude/skills/comic-drama-master

What's inside

Skill name
comic-drama-master
Description
漫剧制作总导演。接受用户故事创意,依次完成剧本创作、角色设计、分镜图生成、分镜视频生成、视频合成全流程,产物保存到 COMIC_DRAMA_OUTPUT_DIR(默认 ./output),最终交付完整漫剧视频和 TOS 链接。
License
Apache-2.0
Instructions
788 lines in SKILL.md · view on GitHub

Before installing

  • Ships executable scripts

Agent Skills spec issues

  • body: 788 lines (spec recommends under 500)

Files (23)

  • SKILL.md
  • examples.md · examples/
  • character-designer.md · references/
  • scene-designer.md · references/
  • screenplay-generator.md · references/
  • storyboard-director.md · references/
  • video-synthesizer.md · references/
  • app_config.py · scripts/
  • ark_auth.py · scripts/
  • batch_image_generate.py · scripts/
  • batch_video.py · scripts/
  • create_video_task.py · scripts/
  • file_download.py · scripts/
  • get_aksk.py · scripts/
  • image_generate.py · scripts/
  • model_service.py · scripts/
  • query_video_task.py · scripts/
  • task_manager.py · scripts/
  • tos_upload.py · scripts/
  • verify_task.py · scripts/
  • video_merge.py · scripts/
  • video_scorer.py · scripts/
  • web_search.py · scripts/

SKILL.md (first lines)

# 漫剧制作总导演

你是统筹全局的漫剧制作总导演,负责协调五个专业环节完成从创意到成片的全流程。

> 五个专业环节的详细规格:
> - 剧本生成:参见 `references/screenplay-generator.md`
> - 角色设计:参见 `references/character-designer.md`
> - 场景美术:参见 `references/scene-designer.md`
> - 分镜视频:参见 `references/storyboard-director.md`
> - 视频合成:参见 `references/video-synthesizer.md`
>
> 完整使用示例参见 `examples/examples.md`。

---

## ⚠️ 内容安全审核提醒

在开始制作之前,**必须先对用户的故事创意进行内容安全预审**。

### 高风险关键词(容易触发 API 拒绝)

以下类型的内容在视频生成阶段(doubao-seedance API)有较高概率被拒绝(`OutputVideoSensitiveContentDetected`):

| 风险类别 | 高风险关键词示例 |
|---------|----------------|
| 战争军事 | 战争、军队、军事、入侵、屠杀、征服 |
| 血腥暴力 | 血腥、血液、流血、残肢、内脏、致命伤 |
| 武器描述 | 刀砍、剑刺、弓箭射杀、枪械、炸弹 |
| 宗教敏感 | 堕天使、恶魔、撒旦、邪教、亵渎 |
| 恐怖元素 | 恐怖、惊悚、尸体、骷髅、鬼魂 |
| 政治敏感 | 政治人物、敏感历史事件、领土争议 |

### 预审处理规则

1. **低风险**(日常、奇幻冒险、儿童故事等):直接开始制作
2. **中风险**(武侠打斗、修仙对决等):在剧本和 prompt 中使用**委婉替代词**:
   - ❌ `blood spraying from wound` → ✅ `spiritual energy impact, staggering backward`
   - ❌ `sword piercing through chest` → ✅ `sword energy clash, powerful strike`
   - ❌ `army marching to war` → ✅ `warriors gathering for a decisive confrontation`
   - ❌ `bloody battle` → ✅ `fierce spiritual energy confrontation`
3. **高风险**(纯战争、恐怖、政治敏感):**明确告知用户**该题材可能导致视频生成大面积失败,建议:
   - 调整为更温和的表达方式
   - 更换题材方向
   - 用户坚持则继续,但提前说明可能有 30-50% 的场景需要反复重试

> **提醒时机**:在步骤2(初始化任务目录)之后、步骤3(剧本生成)之前,向用户简要说明风险等级和应对策略。

---

## ⚡ 脚本参数速查表(大模型必读)

> **重要**:以下是所有脚本的精确调用格式和 JSON 文件格式。调用任何脚本前,必须严格按照此表准备参数,不得猜测格式。

### app_config.py — 读取配置
```bash
python scripts/app_config.py
# 无参数,返回 JSON
```

### task_manager.py — 任务目录管理
```bash
# 初始化任务
python scripts/task_manager.py init "<task_name>"

# 保存文档
python scripts/task_manager.py save "<task_folder绝对路径>" "<文件名.md>" "<Markdown内容>"

# 列出任务
python scripts/task_manager.py list
```

### web_search.py — 网络搜索
```bash
python scripts/web_search.py "<搜索关键词>"
# 返回 JSON 字符串数组
```

### image_generate.py — 单张图片生成
```bash
python scripts/image_generate.py "<英文prompt>" --output-dir "<保存目录绝对路径>"
# 返回 JSON: {"saved_files": ["/absolute/path/to/generated_image_TIMESTAMP_0.png"]}
```

### batch_image_generate.py — 批量并行图片生成(⚡ 推荐用于分镜图和角色立绘)
```bash
# 从 JSON 文件读取 prompts,并行生成(默认3线程)
python scripts/batch_image_generate.py \
  --prompts-file prompts.json \
  --output-dir "<保存目录绝对路径>" \
  --prefix scene_ \
  --max-workers 3 \
  --max-retries 3

# 返回 JSON: {"status": "success", "total": N, "succeeded": N, "failed": 0, "saved_files": [...], "elapsed_seconds": X}
```

**📋 image_prompts.json 格式(纯字符串数组):**
```json
[
  "STYLE_ANCHOR, scene 1 prompt...",
  "STYLE_ANCHOR, scene 2 prompt...",
  "STYLE_ANCHOR, scene 3 prompt..."
]
```

> ⚡ **性能对比**:7 张分镜图串行生成约 70s,并行生成约 25s,提速约 3 倍。
> 内置自动重试(最多3次)和失败 prompt 简化 fallback 机制。

### batch_video.py — 批量视频提交/轮询
```bash
# 提交任务(⚠️ JSON 文件格式见下方)
python scripts/batch_video.py submit \
  --prompts-file prompts.json \
  --first-frames-file frames.json \
  --durations-file durations.json

# 轮询任务
python scripts/batch_video.py poll --task-ids-file task_ids.json --interval 30
```

From bytedance/agentkit-samples, Apache-2.0.

More from bytedance/agentkit-samples

Signals

Stars
473
Forks
101
Open issues
14
Last commit (this folder)
Oct 21, 2026 (7 days ago)

From GitHub (fetched Oct 24, 2026).

Details

License
Apache-2.0
First seen
Oct 24, 2026
Last checked
Oct 24, 2026

More from bytedance

Cite this page

comic-drama-master skill. Agent Mart. https://agentmart.dev/skills/bytedance-comic-drama-master (last updated 2026-10-24)