Skill flagged — suspicious patterns detected

ClawHub Security flagged this skill as suspicious. Review the scan results before using.

AI Animation Studio

v2.1.1

自动将用户故事转分镜、画面和动画生成视频,支持多风格、自动配音、字幕和背景音乐选择。

0· 238·0 current·0 all-time

Install

OpenClaw Prompt Flow

Install with OpenClaw

Best for remote or guided setup. Copy the exact prompt, then paste it into OpenClaw for systiger/ai-animation-studio.

Previewing Install & Setup.
Prompt PreviewInstall & Setup
Install the skill "AI Animation Studio" (systiger/ai-animation-studio) from ClawHub.
Skill page: https://clawhub.ai/systiger/ai-animation-studio
Keep the work scoped to this skill only.
After install, inspect the skill metadata and help me finish setup.
Use only the metadata you can verify from ClawHub; do not invent missing requirements.
Ask before making any broader environment changes.

Command Line

CLI Commands

Use the direct CLI path if you want to install manually and keep every step visible.

OpenClaw CLI

Bare skill slug

openclaw skills install ai-animation-studio

ClawHub CLI

Package manager switcher

npx clawhub@latest install ai-animation-studio
Security Scan
VirusTotalVirusTotal
Benign
View report →
OpenClawOpenClaw
Suspicious
medium confidence
!
Purpose & Capability
The skill's stated purpose (story→storyboard→images→video) matches the code: it calls a doubao media script, ffmpeg, and local resource libraries. However SKILL.md declares an ARK_API_KEY requirement while the registry metadata lists no required env vars, and the code hardcodes absolute paths (C:\Users\10954\.openclaw\workspace\skills\doubao-media\..., D:\AI视频资源\...) tied to the author's machine. Requiring access to another skill's workspace and a D: resources folder is surprising and disproportionate to a reusable skill implementation — a legitimate skill would use relative/installation-aware paths or configurable resource dirs.
!
Instruction Scope
Runtime instructions and code invoke external scripts (doubao_media.py) and ffmpeg (expected for media work) but also scan and read user filesystem resources under D:\AI视频资源 (resource_index.py loads Excel/CSV and enumerates media). That can expose arbitrary filenames and contents in a broad local directory. The SKILL.md and code do not document allowing the user to choose the resource root (it's hardcoded), so the skill may access more of the host filesystem than a user would expect.
Install Mechanism
There is no install spec (instruction-only install), which lowers distribution risk. However the skill package includes Python code that will run on the host. No remote download URLs are used, but the code executes subprocesses (python scripts and ffmpeg) — normal for this domain but worth auditing locally before execution.
!
Credentials
SKILL.md lists ARK_API_KEY (豆包/ARK API) as required, but the registry metadata lists no required environment variables — that's an inconsistency. The code itself does not read generic secrets from env vars, but relies on another skill (doubao-media) and an external API key; giving an API key should be deliberate. Also the skill expects access to the user's D: resource directory and to a doubao-media script in an absolute workspace path — both are broad privileges that should be configurable.
Persistence & Privilege
The skill does not request always:true or other elevated persistent privileges. It does not modify other skills' configs in the provided code. Autonomous invocation is allowed (platform default) but is not combined with other high-risk flags here.
What to consider before installing
What to consider before installing: - Mismatch: SKILL.md says you need ARK_API_KEY (豆包/ARK) but the skill metadata lists no required env vars. Treat any request to provide that API key carefully — only provide it if you trust the doubao/ARK service and the skill. - Hardcoded absolute paths: The code uses C:\Users\10954\.openclaw\workspace\skills\doubao-media\scripts\doubao_media.py and D:\AI视频资源\... — these will try to read your D: drive resources and a specific workspace path. Confirm or change the resource path and the doubao-media script location to avoid unintended file access. - Local filesystem access: resource_index.py scans and loads many files (Excel/CSV/media) under D:\AI视频资源. If you keep sensitive files under that path, the skill could read them. Run the skill in a sandbox or change RESOURCE_DIR to a dedicated folder you control. - Dependencies: The skill depends on doubao-media and edge-tts; inspect those skills before granting permissions. The skill invokes external scripts and ffmpeg via subprocess — audit them, and ensure ffmpeg is the expected binary on PATH. - Copyright/legal risk: BGM list includes recognizable commercial song names. Ensure you have rights to use any music before publishing videos. - Code quality / minor bugs: create_animation.py references random.choice but does not import random at top — indicates limited testing; be prepared for runtime errors. Recommendations: 1) Do not install blindly. Inspect and (if needed) modify the code to replace absolute paths with configurable options. 2) Provide ARK/豆包 API key only if necessary and in a least-privilege manner; prefer scoped credentials or test without keys first. 3) Run in an isolated environment (VM/container) or with a dedicated resource folder to avoid accidental scanning of personal files. 4) Review/verify the doubao-media and edge-tts skills and any third-party binaries (ffmpeg) the skill calls. If you want, I can: (a) point out the exact lines that reference hardcoded paths and the undeclared env var, (b) produce a small patch to make resource dir and doubao script path configurable, or (c) generate a checklist for secure testing.

Like a lobster shell, security has layers — review code before you run it.

latestvk97cdx5cn22j2bjazr0mhqw19h83v53j
238downloads
0stars
2versions
Updated 4w ago
v2.1.1
MIT-0

AI Animation Studio - AI动画制作技能

从故事到动画的全流程自动化制作技能。

版本: 2.1.1
作者: systiger
ClawHub: https://clawhub.ai/systiger/ai-animation-studio

触发词

AI动画、动画制作、生成动画、故事动画、AI视频制作、图生视频

特性

两种生成模式完整工作流程(故事→分镜→画面→动画→配音→BGM→合成) ✅ 皮克斯/宫崎骏/童话等多种风格自动配音+字幕BGM背景音乐选择(30+首热门BGM)⭐ NEW竖版/横版自适应

快速开始

# 模式一:图片转视频(速度快)
python create_animation.py --mode 1 --output "D:/output"

# 模式二:图片动画化再生成视频(效果更强)
python create_animation.py --mode 2 --output "D:/output"

# 只生成指定场景
python create_animation.py --mode 2 --scene 3

工作流程

第一步:需求收集

用户输入故事后,自动提问:

  1. 生成模式

    • A. 图片转视频(速度快,保留原画面细节)
    • B. 图片动画化再生成视频(动画效果更强)
  2. 风格选择

    • 影视风格:王家卫、诺兰、宫崎骏、漫威等
    • 动画风格:皮克斯、卡通、手绘等
    • 纪录片风格:BBC、Netflix等
  3. 视频规格

    • 竖版(9:16)- 短视频平台
    • 横版(16:9)- YouTube/B站
    • 方形(1:1)- 社交媒体
  4. BGM背景音乐选择 🎵

    • A. 自动推荐 - AI根据故事风格自动选择
    • B. 从BGM库选择 - 查看30+首热门BGM列表
    • C. 随机选择 - 从推荐分类随机抽取
    • D. 无BGM - 仅保留配音

    BGM分类

    • 🌸 儿童动画:菊次郎的夏天、DJ喜羊羊、嘟嘟嘟嘟哒哒哒
    • 🔥 励志热血:Beautiful Now、The Spectre、易燃易爆炸
    • 😂 搞笑娱乐:喜剧背景音效、滑稽背景音乐、爱滴魔力转圈圈
    • 💕 浪漫温馨:恋爱循环、Hold On、雨爱
    • 🚀 科幻冒险:The Spectre、Geisha、Run Free

第二步:脚本分解

基于分镜模板,将故事分解为:

场景镜头动作开头方式结尾方式画面风格
场景1镜头描述开场介绍呼吁互动风格描述

第三步:画面生成

使用豆包API生成每个分镜画面:

  • 人物一致性(固定角色形象)
  • 风格统一性(同一画风)
  • 场景连贯性(背景协调)

第四步:动画生成(核心)⭐

两种生成模式

🎬 模式一:图片转视频(推荐)

直接用生成的图片转成视频,保留原画面细节:

python doubao_media.py vid "镜头缓慢推进,人物轻微动作" --image "图片URL" --ratio 9:16 --duration 5

特点:速度快、画质高、适合静态场景

🎨 模式二:图片动画化再生成视频

先生成动画风格的图片,再添加动态效果:

# 第一步:生成动画风格图片
python doubao_media.py img "动画风格提示词" --size 1024x1792

# 第二步:图片转视频
python doubao_media.py vid "动画效果描述" --image "图片URL" --duration 5

特点:动画效果更强、角色动作更丰富、适合人物场景

动态效果类型

  • 镜头运动:推/拉/摇/移/环绕
  • 角色动画:说话/眨眼/手势/走动
  • 场景动效:风吹/水流/光影/粒子

第五步:配音字幕

  • 根据角色分配不同音色
  • 自动生成字幕时间轴
  • 字幕样式与视频风格匹配

第六步:特效合成

  • 添加特效素材(转场/叠加/滤镜)
  • 添加音效(环境音/特效音)
  • 合并所有场景视频
  • 添加配音
  • 最后添加BGM背景音乐(BGM音量30%,配音音量100%):
    # 工作流程:
    # 1. 场景1.mp4 + 场景2.mp4 + ... → 合并视频.mp4
    # 2. 合并视频.mp4 + 配音.mp3 → 视频_配音.mp4
    # 3. 视频_配音.mp4 + BGM.mp3 → 最终视频.mp4
    
    ffmpeg -i video_with_voice.mp4 -i bgm.mp3 -filter_complex \
      "[1:a]volume=0.3,aloop=loop=-1:size=2e+09[bgm];[0:a][bgm]amix=inputs=2:duration=first:dropout_transition=2[aout]" \
      -map 0:v -map "[aout]" -c:v copy final_video.mp4
    

资源库索引

分镜模板

文件名用途
AI视频脚本分镜模板_共300条.xlsx场景/镜头/开头结尾/画面风格
【先看这个】分镜画面提示词.xlsx镜头类型提示词
即梦100多组(700+个)神级指令合集.xlsxAI生图指令模板
300+电影风格提示词.csv电影风格参考

特效素材

类别数量用途
划屏元素素材24个转场特效
竖版动态背景86个背景叠加
点赞心形素材21个互动特效
唯美炫光素材15个光效叠加

音效素材

类别用途
卡点音乐节奏背景
字幕弹出音效文字动画
网红爆款bgm流行配乐

使用示例

用户:帮我做一个关于"森林里的小精灵"的动画视频

AI:
好的,我来帮你制作"森林里的小精灵"动画视频。

📋 首先确认几个问题:

1️⃣ 生成模式:
   A. 图片转视频(速度快,保留原画面细节)
   B. 图片动画化再生成视频(动画效果更强)

2️⃣ 风格选择:
   A. 宫崎骏动画风(自然色彩+清新构图)
   B. 皮克斯动画质感(明亮色彩+卡通建模)
   C. 童话故事风(柔和色调+手绘质感)

3️⃣ 视频规格:
   A. 竖版 9:16(抖音/快手)
   B. 横版 16:9(B站/YouTube)

4️⃣ BGM背景音乐:
   A. 自动推荐(AI根据风格选择)
   B. 从BGM库选择(查看30+首BGM列表)
   C. 随机选择
   D. 无BGM

请回复选项,或让我自动选择推荐配置。

BGM选择示例

AI:🎵 为您的"森林里的小精灵"动画推荐以下BGM:

🌸 儿童动画风格推荐:
  1. 菊次郎的夏天 - 温馨童年氛围
  2. DJ喜羊羊 - 活泼欢快节奏
  3. 嘟嘟~嘟嘟~哒哒哒 - 可爱童趣旋律

请选择BGM编号,或回复"随机"让AI选择。

成功案例

《小小科学家的神奇发明》

  • 模式:模式二(图片动画化)
  • 时长:31.79秒
  • 场景:7个动态场景
  • 风格:皮克斯动画
  • 配音:4个角色
  • 文件大小:8.6MB

技术实现

依赖技能

  • doubao-media - AI生图/生视频
  • edge-tts - 配音生成
  • ffmpeg - 视频合成

API要求

输出格式

  • MP4(H.264 + AAC)
  • 分辨率:1088x1920(竖版)/ 1920x1080(横版)
  • 帧率:24fps
  • 字幕:内嵌烧录

文件结构

ai-animation-studio/
├── SKILL.md              # 技能说明
├── create_animation.py   # 主脚本(两种模式)
├── resource_index.py     # 资源索引
├── workflow.py           # 工作流
└── scripts/              # 示例脚本
    └── xiaoming_scientist.json

更新日志

v2.1.1 (2026-03-29)

  • ✅ 优化BGM合成时机:在最后阶段才添加BGM
  • ✅ 改进工作流程:场景合并 → 配音 → BGM(三步流程)
  • ✅ 添加BGM循环功能,自动匹配视频时长
  • ✅ 优化BGM音量混合(默认30%)

v2.1.0 (2026-03-29)

  • ✅ 新增BGM背景音乐选择功能
  • ✅ 支持5大分类(儿童动画/励志热血/搞笑娱乐/浪漫温馨/科幻冒险)
  • ✅ 支持自动推荐、随机选择、手动选择
  • ✅ BGM音量可调节(默认30%)
  • ✅ 新增30+首热门BGM资源库

v2.0.0 (2026-03-29)

  • ✅ 新增两种生成模式
  • ✅ 完整的图生视频流程
  • ✅ 自动配音合成
  • ✅ 成功案例:小小科学家

v1.0.0 (2026-03-29)

  • 初始版本
  • 基础工作流程

创建时间:2026-03-29
_资源位置:D:\AI视频资源_
作者:systiger

Comments

Loading comments...