← 返回 Skills 市场
858
总下载
2
收藏
3
当前安装
1
版本数
在 OpenClaw 中安装
/install youtube-knowledge-extractor
功能描述
Multimodal YouTube video analysis through both audio (transcript) and visual (frame extraction + image analysis) channels. Especially powerful for HowTo vide...
安全使用建议
This skill appears coherent and limited to downloading and analyzing YouTube content. Before installing, confirm you are comfortable with: (1) the skill downloading videos and writing temporary files to /tmp (potentially large disk/network usage), (2) network access to YouTube/captions endpoints, and (3) the source used to install yt-dlp — verify the 'uv' installer or install yt-dlp from a trusted package source yourself. Also consider copyright and terms-of-service implications for downloading videos and transcripts. If you need stricter controls, run the skill in an isolated environment or container.
功能分析
Type: OpenClaw Skill
Name: youtube-knowledge-extractor
Version: 1.1.0
The OpenClaw skill 'youtube-knowledge-extractor' is classified as benign. All operations, including the use of `yt-dlp`, `ffmpeg`, `ffprobe`, `curl`, and embedded Python scripts, are directly aligned with its stated purpose of multimodal YouTube video analysis. The skill creates and cleans up temporary files in `/tmp`, and network requests are made to YouTube-related endpoints for video and subtitle data. There is no evidence of prompt injection attempting to subvert the agent's purpose, exfiltrate data, establish persistence, or perform other malicious actions. The instructions to the agent are detailed but focus on effective task execution within the skill's scope.
能力评估
Purpose & Capability
Name/description (multimodal YouTube analysis) aligns with required binaries (yt-dlp, ffmpeg, python3, curl) and the described steps (download metadata/subtitles, download video, extract frames, analyze). The declared install (yt-dlp) is appropriate for the task.
Instruction Scope
SKILL.md contains explicit shell/python commands that operate only on a temp working directory, fetch YouTube metadata/subtitles via yt-dlp/curl, download the video, and run ffmpeg/ffprobe. It does not instruct reading unrelated system files, environment variables, or exfiltrating data to unexpected endpoints.
Install Mechanism
Install uses a 'uv' package entry to provide yt-dlp. Installing yt-dlp is expected, but 'uv' is an uncommon installer in this metadata — user may want to confirm the installer/source provenance for yt-dlp on their platform.
Credentials
No environment variables, credentials, or config paths are requested. Network access to YouTube/captions endpoints is required and consistent with the skill's purpose.
Persistence & Privilege
Skill is not always-enabled, does not request elevated privileges, and only writes artifacts to a temporary working directory under /tmp. It does not alter system or other-skill configurations.
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 youtube-knowledge-extractor - 安装完成后，直接呼叫该 Skill 的名称或使用
/youtube-knowledge-extractor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
v1.1.0
Renamed from youtube-video-analyzer to youtube-knowledge-extractor for better discoverability
元数据
常见问题
Youtube Knowledge Extractor 是什么？
Multimodal YouTube video analysis through both audio (transcript) and visual (frame extraction + image analysis) channels. Especially powerful for HowTo vide... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 858 次。
如何安装 Youtube Knowledge Extractor？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install youtube-knowledge-extractor」即可一键安装，无需额外配置。
Youtube Knowledge Extractor 是免费的吗？
是的，Youtube Knowledge Extractor 完全免费（开源免费），可自由下载、安装和使用。
Youtube Knowledge Extractor 支持哪些平台？
Youtube Knowledge Extractor 跨平台运行，可在任意部署了 OpenClaw / Claude Code 的环境中使用（linux, macos）。
谁开发了 Youtube Knowledge Extractor？
由 sdrabent（@sdrabent）开发并维护，当前版本 v1.1.0。
推荐 Skills