← 返回 Skills 市场
1264
总下载
2
收藏
3
当前安装
2
版本数
在 OpenClaw 中安装
/install doubao-video-analyzer
功能描述
使用豆包2.0模型解析视频。当需要执行分析视频内容等需要理解视频视觉信息时调用该技能。你必须在持有本地视频路径或网络视频链接时才能调用该技能
使用说明 (SKILL.md)
\r \r
Doubao Video Analyze\r
\r
隐私与数据安全声明\r
\r 重要:本技能通过火山引擎 Ark API(https://ark.cn-beijing.volces.com)处理视频内容。使用此技能时:\r \r
- 本地视频文件将被编码为 base64 并上传至火山引擎服务器\r
- 远程视频 URL 将直接发送给火山引擎服务器进行解析\r
- 请确保视频内容不包含敏感、机密或未经授权的信息\r
- 视频数据的处理遵循火山引擎的服务条款和隐私政策\r \r
使用方法\r
\r
注意\r
\r 对于远程网络视频,如果通过该技能无法执行,请直接通知用户错误内容,永远不要尝试使用web_fetch之类的工具去重试!\r \r
前置检查\r
\r
- 检查当前是否有
ARK_API_KEY环境变量,如果没有,直接终止并引导用户先进行配置\r - 直接尝试调用如下bash指令,确保用户当前python环境有openai依赖\r \r
pip install 'volcengine-python-sdk[ark]'\r
```\r
\r
### 分析视频\r
\r
1. 你需要先明确你的分析目的是什么,你需要自行写出一段提示词,例如\r
\r
```plaintext\r
### 角色\r
你是一位xxxxx,善于分析视频的xxxxxx\r
\r
### 任务\r
你的任务是xxxxxxx\r
\r
### 输出\r
你需要输出以下内容:\r
xxxxxxxxxxxxxx\r
```\r
\r
**注意**:你需要写出高质量的提示词,但并不一定要按照上面的模板\r
\r
2. 调用本技能目录下脚本以获取分析结果\r
\r
- 本地视频\r
\r
```bash\r
python script/parse_video.py --video 'absolute/video/path.mp4' --prompt 'your prompt content'\r
```\r
\r
\r
\r
- 远程视频链接\r
\r
```bash\r
python script/parse_video.py --video 'https://example.com/video_url.mp4' --prompt 'your prompt content' --remote\r
```\r
\r
\r
\r
\r
等待该命令执行结束并获取结果作为该视频的分析结果\r
\r
\r
\r
### 网络视频错误处理\r
\r
如果遇到网络视频超过大小限制的问题,首先现将该网络视频下载至你的workspace的video目录下,然后再通过本地视频的方式调用
安全使用建议
This skill will send your video data (local files uploaded via the SDK or remote URLs forwarded) to Volcengine's Ark service and requires your ARK_API_KEY. Only use it with non-sensitive or permissible content, and be sure you trust the ARK endpoint and your API key. Installing the recommended Python SDK will pull code from PyPI — review that package if you need to. Keep the skill user-invocable (not always-on) and do not share your ARK_API_KEY with untrusted parties. If you need assurance on data retention/privacy, review Volcengine's service and privacy terms before uploading videos.
功能分析
Type: OpenClaw Skill
Name: doubao-video-analyzer
Version: 1.0.1
The skill is a legitimate integration for analyzing video content using the Doubao 2.0 model via the Volcengine (火山引擎) Ark API. The Python script `script/parse_video.py` correctly implements file uploading and API interaction with the official endpoint (ark.cn-beijing.volces.com), and the `SKILL.md` provides transparent instructions regarding privacy, data handling, and dependency installation. No evidence of malicious intent, data exfiltration, or unauthorized execution was found.
能力评估
Purpose & Capability
Name/description, declared requirements (python + ARK_API_KEY), and the included script all align: the skill uploads or references a video to Volcengine Ark and requests analysis from the doubao model. No unrelated credentials or binaries are requested.
Instruction Scope
SKILL.md and the script are scoped to the stated task: check ARK_API_KEY, install the Volcengine SDK, craft a prompt, and run the provided script to either upload a local file or pass a remote URL. Note: SKILL.md mentions base64-encoding local files for upload while the script uses the SDK's file upload; this is an implementation detail rather than a scope expansion. The doc explicitly warns against using web_fetch to retry remote URLs, which restricts behavior.
Install Mechanism
There is no formal install spec (instruction-only), but SKILL.md instructs the user to pip install 'volcengine-python-sdk[ark]'. Installing a third-party Python package from PyPI is expected for this functionality but does cause network/third-party code to be installed; the package name appears to match the provider and there are no mysterious download URLs.
Credentials
Only ARK_API_KEY is required and used by the script. That single credential is proportional to the skill's need to authenticate to the Ark API. Users should understand this key authorizes uploading video data to the vendor.
Persistence & Privilege
The skill is not always-enabled and does not request elevated persistence or modify other skills or system settings. It runs only when invoked.
如何使用
- 确保已安装 OpenClaw(本地或 Docker 部署)
- 在对话框中输入安装命令:
/install doubao-video-analyzer - 安装完成后,直接呼叫该 Skill 的名称或使用
/doubao-video-analyzer触发 - 根据 Skill 的参数说明提供必要输入,即可获得结构化输出
版本历史
v1.0.1
- Added a privacy and data security statement, clarifying that videos are uploaded to and processed by Volcengine Ark API and outlining handling of local and remote videos.
- Embedded metadata describing required binaries, environment variables, and primary environment variable (ARK_API_KEY).
- No functional or usage changes to the video analysis workflow.
- Documentation updated to ensure user awareness of privacy considerations and data processing locations.
v1.0.0
- Initial release of doubao-video-analyze skill.
- Enables video content analysis using the Doubao 2.0 model with local file paths or network video links.
- Provides step-by-step usage instructions, including prerequisite checks for environment variables and dependencies.
- Outlines prompt creation for effective video analysis.
- Details command-line usage for both local and remote video files.
- Includes clear guidance on handling errors, especially with remote video size limits.
元数据
常见问题
Video analyze by doubao2.0 是什么?
使用豆包2.0模型解析视频。当需要执行分析视频内容等需要理解视频视觉信息时调用该技能。你必须在持有本地视频路径或网络视频链接时才能调用该技能. 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件,目前累计下载 1264 次。
如何安装 Video analyze by doubao2.0?
在 OpenClaw 或 Claude Code 对话框中运行命令「/install doubao-video-analyzer」即可一键安装,无需额外配置。
Video analyze by doubao2.0 是免费的吗?
是的,Video analyze by doubao2.0 完全免费,采用 MIT-0 许可证,可自由下载、安装和使用。
Video analyze by doubao2.0 支持哪些平台?
Video analyze by doubao2.0 跨平台运行,可在任意部署了 OpenClaw / Claude Code 的环境中使用(cross-platform)。
谁开发了 Video analyze by doubao2.0?
由 danielcy(@danielcy)开发并维护,当前版本 v1.0.1。
推荐 Skills