← 返回 Skills 市场
141
总下载
0
收藏
0
当前安装
1
版本数
在 OpenClaw 中安装
/install kai-youtube
功能描述
Download and transcribe YouTube videos using yt-dlp and Whisper CLI, saving audio and transcripts for playback and summary from any YouTube URL.
使用说明 (SKILL.md)
Kai YouTube Skill
Watch YouTube videos by downloading audio and transcribing with Whisper.
Organization
- Download folder:
/home/kai/.openclaw/workspace/kai-yt-videos/ - Audio files:
kai-yt-videos/kai_yt_${VIDEO_ID}.mp3 - Transcripts:
kai-yt-videos/kai_yt_${VIDEO_ID}.txt
Workflow
- Download audio using yt-dlp:
yt-dlp --extract-audio --audio-format mp3 --output "{WORKSPACE}/kai_yt_${VIDEO_ID}.mp3" "\x3CURL>"
- Transcribe using Whisper CLI:
whisper "{WORKSPACE}/kai_yt_${VIDEO_ID}.mp3" --model base --output_format txt --output_dir "{WORKSPACE}"
- Read transcript from
{WORKSPACE}/kai_yt_${VIDEO_ID}.txt
Requirements
yt-dlp- YouTube audio downloader (brew install yt-dlp)whisper- OpenAI Whisper CLI (brew install openai-whisper)
Usage
bash {baseDir}/scripts/youtube.sh "\x3CYouTube_URL>" [--language \x3Clang>]
Tips
- Each video gets unique files (no caching issues)
- Long videos may take several minutes to transcribe
- Language auto-detected if not specified
- Add
--language \x3Clang>for specific language (e.g.,--language Spanish) - All files organized in
kai-yt-videos/folder
Video ID Extraction
Handles both URL formats:
https://www.youtube.com/watch?v=VIDEO_IDhttps://youtu.be/VIDEO_ID
安全使用建议
This skill appears to do what it says: it will download audio from a YouTube URL using yt-dlp and transcribe it locally with the Whisper CLI, saving files under /home/kai/.openclaw/workspace/kai-yt-videos. Before installing/running: (1) ensure you trust the source of the script (owner unknown) — running scripts from unknown authors has risk; (2) install and verify yt-dlp and the Whisper CLI yourself (the skill will not install them); (3) note the script hard-codes the workspace path to /home/kai — edit it to use your agent's workspace or a path you control to avoid permission or path mismatches; (4) be aware content is downloaded locally (copyright/privacy considerations) and transcription may be CPU/GPU intensive; (5) if you want stronger guarantees, inspect or run the script in a sandbox first. If you want, I can suggest a safer variant that uses a non-hardcoded workspace and basic input validation.
功能分析
Type: OpenClaw Skill
Name: kai-youtube
Version: 1.0.0
The skill provides legitimate functionality for downloading YouTube audio and generating transcripts using 'yt-dlp' and 'whisper' CLI tools. The bash script 'scripts/youtube.sh' correctly implements the workflow described in 'SKILL.md', using standard paths and parameters consistent with its stated purpose without any signs of malicious intent or data exfiltration.
能力评估
Purpose & Capability
Name/description, SKILL.md, and the included script all consistently implement downloading audio with yt-dlp and transcribing with the Whisper CLI. Required binaries (yt-dlp, whisper) are appropriate. Minor oddity: the script hard-codes WORKSPACE=/home/kai/.openclaw/workspace/kai-yt-videos which may not match the agent environment or desired workspace and could cause permission or path surprises; otherwise there are no unrelated requirements.
Instruction Scope
Runtime instructions (SKILL.md) and the script limit actions to extracting a YouTube video ID, downloading audio via yt-dlp, transcribing with whisper, writing files to the workspace, and printing the transcript. The skill does not read unrelated files or environment variables, nor does it contact external endpoints other than what yt-dlp uses to fetch YouTube content.
Install Mechanism
No install spec is provided (instruction-only plus a shell script). This is lower risk because nothing is silently downloaded/installed by the skill itself. It does rely on the user/system to have yt-dlp and whisper installed.
Credentials
The skill requests no environment variables or credentials. It writes files to a local folder (hard-coded path under /home/kai). No secret access or unrelated credentials are requested.
Persistence & Privilege
Skill does not request persistent platform privileges, does not set always:true, and does not modify other skills or system config. It simply writes audio and transcript files under the workspace directory.
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 kai-youtube - 安装完成后，直接呼叫该 Skill 的名称或使用
/kai-youtube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
v1.0.0
- Initial release of kai-youtube skill.
- Watch and transcribe YouTube videos by downloading audio with yt-dlp and transcribing via Whisper CLI.
- Organizes audio and transcripts by video ID in the workspace folder.
- Supports any YouTube URL.
- Includes trigger phrases for easy activation.
元数据
常见问题
Kai YouTube 是什么？
Download and transcribe YouTube videos using yt-dlp and Whisper CLI, saving audio and transcripts for playback and summary from any YouTube URL. 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 141 次。
如何安装 Kai YouTube？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install kai-youtube」即可一键安装，无需额外配置。
Kai YouTube 是免费的吗？
是的，Kai YouTube 完全免费，采用 MIT-0 许可证，可自由下载、安装和使用。
Kai YouTube 支持哪些平台？
Kai YouTube 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 Kai YouTube？
由 Blaze🔥（@ogdegenblaze）开发并维护，当前版本 v1.0.0。
推荐 Skills