Skill flagged — suspicious patterns detected

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

Seedance AI Video Gen

v1.0.0

使用火山引擎 Seedance 2.0 系列模型(doubao-seedance-2-0 / doubao-seedance-2-0-fast)通过方舟平台 API 生成高质量 AI 视频。支持文生视频、图生视频、视频参考、音频参考等多模态内容生成,适用于用户要求生成/制作/创建视频、文生视频、图生视频、AI 生...

0· 82·0 current·0 all-time
byCong Pendy@jancong

Install

OpenClaw Prompt Flow

Install with OpenClaw

Best for remote or guided setup. Copy the exact prompt, then paste it into OpenClaw for jancong/jancong-seedance-video-gen.

Previewing Install & Setup.
Prompt PreviewInstall & Setup
Install the skill "Seedance AI Video Gen" (jancong/jancong-seedance-video-gen) from ClawHub.
Skill page: https://clawhub.ai/jancong/jancong-seedance-video-gen
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 jancong-seedance-video-gen

ClawHub CLI

Package manager switcher

npx clawhub@latest install jancong-seedance-video-gen
Security Scan
VirusTotalVirusTotal
Benign
View report →
OpenClawOpenClaw
Suspicious
high confidence
Purpose & Capability
The skill's name/description match the included scripts and instructions: it submits generation tasks to the Ark/Seedance API and downloads results. However the registry metadata lists no required environment variables or primary credential while the SKILL.md and all scripts require ARK_API_KEY (and optionally ARK_API_URL). That metadata omission is inconsistent and should be corrected.
Instruction Scope
SKILL.md and scripts only perform actions related to submitting tasks, polling status, and downloading results. They require user-provided public URLs for reference assets and instruct uploading local files externally if needed. The instructions do not ask to read unrelated files or exfiltrate other system data.
Install Mechanism
No install spec or external downloads are used; the skill is instruction-only with bundled Python scripts. No third-party package downloads or archive extraction are present in the manifest.
!
Credentials
The scripts legitimately need an API credential (ARK_API_KEY) and optionally ARK_API_URL. The manifest/registry metadata incorrectly states there are no required env vars or primary credential. That mismatch is concerning because users may not be warned that they must supply an API key. Otherwise no unrelated credentials or excessive env access are requested.
Persistence & Privilege
The skill does not request permanent 'always' inclusion and does not modify other skills or system-wide configs. It runs as a user-invoked skill and the scripts perform only network requests to the configured Ark API and to user-supplied asset URLs.
What to consider before installing
This skill appears to implement Seedance/Ark video generation, but the package metadata fails to declare the required ARK_API_KEY (and ARK_API_URL is configurable). Before installing: 1) Treat the source as untrusted until you verify the publisher (homepage is missing and owner is unknown). 2) Do not set an API key with broad or production privileges until you confirm which Ark account will be used; prefer a scoped/test key. 3) Be aware the scripts will make outbound HTTPS requests to the configured ARK_API_URL and to any user-provided asset URLs (images/videos/audio). 4) If you need guarantees, request the maintainer to update registry metadata to list ARK_API_KEY as a required credential and provide provenance (homepage/repo). If you cannot verify the source, consider alternative, well-known integrations or inspect/execute the scripts in a sandboxed environment.

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

latestvk972rb2rr279sv8j3ys9eqrpvd84kmhk
82downloads
0stars
1versions
Updated 2w ago
v1.0.0
MIT-0

Seedance 视频生成 Skill

使用火山引擎方舟平台的 Seedance 2.0 系列 AI 模型生成高质量视频。

支持的模型

模型 ID特点适用场景
doubao-seedance-2-0-260128标准版,质量更高高质量商业视频、精细画面
doubao-seedance-2-0-fast-260128快速版,速度更快快速预览、批量测试

工作流程

Step 1: 准备请求参数

根据用户需求构建 API 请求体:

必需参数:

  • model: 模型名称(从上表选择)
  • content: 多模态内容数组,至少包含一个 text 类型元素

可选参数:

  • generate_audio: 是否同时生成音频/语音 (true/false)
  • ratio: 视频比例 "16:9" | "9:16" | "1:1"(默认 16:9)
  • duration: 视频时长秒数,范围 5-11(默认 10)
  • watermark: 是否添加水印 (true/false)

Step 2: 构建 content 数组

content 数组支持以下类型的多模态输入:

{
  "content": [
    // 1. 文本提示词(必需)
    {
      "type": "text",
      "text": "详细的视频描述提示词..."
    },

    // 2. 参考图片(可选,可多个)
    {
      "type": "image_url",
      "image_url": { "url": "https://example.com/image.jpg" },
      "role": "reference_image"
    },

    // 3. 参考视频(可选,用于参考运镜/构图风格)
    {
      "type": "video_url",
      "video_url": { "url": "https://example.com/video.mp4" },
      "role": "reference_video"
    },

    // 4. 参考音频/背景音乐(可选)
    {
      "type": "audio_url",
      "audio_url": { "url": "https://example.com/audio.mp3" },
      "role": "reference_audio"
    }
  ]
}

注意: 图片、视频、音频 URL 必须是公网可访问的地址。如果用户提供本地文件,需要先上传到 OSS 或使用 Base64 编码。

Step 3: 提交生成任务

调用 scripts/generate_video.py 脚本提交异步任务:

python3 ~/.workbuddy/skills/seedance-video-gen/scripts/generate_video.py \
  --model doubao-seedance-2-0-fast-260128 \
  --prompt "你的视频描述" \
  --ratio 16:9 \
  --duration 10 \
  [--image-url "https://..."] \
  [--video-url "https://..."] \
  [--audio-url "https://..."] \
  [--generate-audio] \
  [--no-watermark]

或者直接使用 curl 调用 API(见 references/api_reference.md)。

API 返回任务 ID:

{ "id": "cgt-20260410162021-xxxxx" }

Step 4: 轮询任务状态

提交后任务为异步处理,需轮询查询状态:

python3 ~/.workbuddy/skills/seedance-video-gen/scripts/query_task.py <task_id>

或直接 curl:

curl -X GET "{ARK_API_URL}/api/v3/contents/generations/tasks/{task_id}" \
  -H "Authorization: Bearer {API_KEY}"

状态值:

状态含义操作
pending排队中继续等待
running生成中继续等待
completed完成获取结果
failed失败查看错误信息

完成时的响应包含:

  • video_result.video_url: 生成的视频地址
  • video_result.cover_image_url: 封面图地址

Step 5: 下载并展示结果

任务完成后:

  1. 下载视频文件到工作目录
  2. 使用 preview_url 或 open_result_view 展示给用户
  3. 告知用户视频路径和基本信息

提示词撰写建议

参考 references/prompt_guide.md 获取完整的提示词撰写指南。核心原则:

  1. 明确主体和动作: 描述清楚谁在做什么
  2. 指定镜头语言: 运镜方式(推拉摇移)、视角(第一人称/俯拍等)
  3. 时间轴描述: 用时间戳分段描述不同时段的画面变化,如 0-2 秒:xxx;3-5 秒:xxx
  4. 风格关键词: 明确视觉风格(写实/动漫/电影感等)
  5. 音效描述: 如果 generate_audio=true,可在提示词中描述音效和配音需求
  6. 参考素材说明: 当有图片/视频参考时,在提示词中指明如何使用这些素材

环境配置

脚本从环境变量读取以下配置:

环境变量说明默认值
ARK_API_KEY方舟平台 API Key(必填)
ARK_API_URL方舟 API 地址https://ark.cn-beijing.volces.com

确保在使用前已设置 ARK_API_KEY 环境变量。

Comments

Loading comments...