菜单
This skill should be used when the user asks to "download YouTube transcript", "get YouTube captions", "extract YouTube subtitles", or needs to retrieve transcript from a YouTube video.
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
This skill should be used when the user asks to "create an agent", "add an agent", "write a subagent", "agent frontmatter", "when to use description", "agent examples", "agent tools", "agent modes", "primary agent", "autonomous agent", or needs guidance on agent structure, system prompts, triggering conditions, or agent development best practices for OpenCode.
This skill should be used when the user asks to "create a 3D model with build123d", "design parametric parts", "use builder mode", "use algebra mode", "select faces or edges", "export STEP or STL", or works with build123d CAD programming.
This skill should be used when the user asks to "create a slash command", "add a command", "write a custom command", "define command arguments", "use command frontmatter", "organize commands", "create command with file references", "interactive command", "use AskUserQuestion in command", or needs guidance on slash command structure, YAML frontmatter fields, dynamic arguments, bash execution in commands, user interaction patterns, or command development best practices for OpenCode.
This skill should be used when the user asks to "create a memory", "read memories", "manage memories", "document patterns", or when starting work on a Serena project with existing memories. Essential for maintaining project knowledge across conversation sessions and distinguishing documentation from agent-focused memories.
This skill should be used when the user asks to "find a symbol", "locate code", "refactor", "rename across codebase", or performs ANY coding task with Serena toolbox. Use Language Server Protocol (LSP) for semantic navigation, symbol discovery, finding references, and intelligent editing. Applies to ALL coding tasks - dramatically improves code quality and reduces token usage.
This skill should be used when the user asks to "perform complex task", "implement advanced pattern", "handle multiple scenarios", or needs comprehensive guidance with validation tools. Demonstrates full-featured skill structure.
| name | youtube-transcribe |
| description | This skill should be used when the user asks to "download YouTube transcript", "get YouTube captions", "extract YouTube subtitles", or needs to retrieve transcript from a YouTube video. |
| license | MIT |
| compatibility | opencode |
| metadata | {"category":"media","version":"2.0.0"} |
Use this skill when:
This skill guides downloading YouTube transcripts using yt-dlp and formatting them as structured HTML.
Core principle: Create HTML transcript with title, chapter headings (using video chapters if available), and paragraph content. Abort cleanly if transcript unavailable.
Create one file: transcript.html - Structured HTML with:
<h1> title from video<h2> chapter headings with start timestamp only (format: MM:SS - Title)<p> paragraph content under each chapterChapter sources (in priority order):
| Task | Command | Notes |
|---|---|---|
| Check transcript availability | yt-dlp --list-subs URL | Run first to verify |
| Get video metadata + chapters | yt-dlp --print "%(title)s|%(chapters)s" URL | Chapters may be null |
| Download transcript | yt-dlp --write-auto-sub --sub-lang en --sub-format vtt --skip-download URL | VTT has timestamps |
| Download manual subs (fallback) | yt-dlp --write-sub --sub-lang en --sub-format vtt --skip-download URL | If auto unavailable |
yt-dlp --list-subs VIDEO_URL
If no subtitles available:
yt-dlp --print "%(title)s|%(description)s" VIDEO_URL
Check description for chapter markers:
MM:SS Title or HH:MM:SS Title on separate lines0:00 Introduction
2:30 Main Topic
5:15 Conclusion
If chapters exist in description:
MM:SS - Title (start timestamp and title only)If no chapters in description:
# Try auto-generated first
yt-dlp --write-auto-sub --sub-lang en --sub-format vtt --skip-download VIDEO_URL
# If auto unavailable, try manual
yt-dlp --write-sub --sub-lang en --sub-format vtt --skip-download VIDEO_URL
Parse VTT file:
<00:00:00.000>, <c>, etc.00:00:00.000 --> 00:00:00.000If using video chapters (from description):
MM:SS - Chapter TitleIf creating auto-generated chapters:
MM:SS - Generated Title<!DOCTYPE html>
<html>
<head>
<meta charset="UTF-8">
<title>Video Title Here</title>
</head>
<body>
<h1>Video Title Here</h1>
<h2>0:00 - Introduction</h2>
<p>First paragraph of this chapter...</p>
<p>Second paragraph of this chapter...</p>
<h2>2:30 - Main Topic</h2>
<p>Content for this section...</p>
</body>
</html>
Requirements:
<h1> with video title<h2> with format MM:SS - TitleRemove temporary VTT file after processing.
Wrong: Always creating auto-generated 1-3 minute chapters even when video has chapter markers in description.
Right:
Wrong (no chapters):
<body>
<h1>Video Title</h1>
<p>All transcript text...</p>
</body>
Right:
<body>
<h1>Video Title</h1>
<h2>0:00 - Introduction</h2>
<p>Transcript text for this chapter...</p>
<h2>2:30 - Next Topic</h2>
<p>More content...</p>
</body>
Wrong:
<h2>Introduction</h2>
<h2>0:00 Introduction</h2>
<h2>00:00:00.000 - Introduction</h2>
<h2>0:00 - 2:30 - Introduction</h2>
Right:
<h2>0:00 - Introduction</h2>
Wrong:
webfetch https://www.youtube.com/watch?v=...
Right:
yt-dlp --write-auto-sub --sub-format vtt --skip-download URL
Wrong (using first sentence verbatim):
<h2>0:00 - Can you break down the difference between capitalist, socialist, commu</h2>
Right (concise summary):
<h2>0:00 - Economic Systems Overview</h2>
Guidelines for auto-generated titles:
If you catch yourself thinking:
<h1> with video title.<h2> chapters.MM:SS - Title.1. User requests YouTube transcript
↓
2. Check transcript availability (yt-dlp --list-subs)
↓
3a. If unavailable → Abort with message
3b. If available → Continue
↓
4. Get video title and description
↓
5. Check description for chapter markers
↓
6a. Chapters found → Use those timestamps/titles
6b. No chapters → Generate 1-3 min sections
↓
7. Download transcript (VTT format)
↓
8. Process VTT: remove tags, extract timestamps, clean text
↓
9. Group content by chapters
↓
10. Create transcript.html:
- <h1> with title
- <h2> chapters with MM:SS - Title
- <p> paragraphs under each chapter
↓
11. Clean up temporary files
Video with chapters in description:
<!DOCTYPE html>
<html>
<head>
<meta charset="UTF-8">
<title>Understanding Economic Systems</title>
</head>
<body>
<h1>Understanding Economic Systems</h1>
<h2>0:00 - Introduction</h2>
<p>Welcome to this video on economic systems. Today we'll explore the key differences...</p>
<h2>2:30 - Capitalism Explained</h2>
<p>Capitalism is a system where capital goes to those who can generate the best returns...</p>
<h2>5:45 - Socialist Approaches</h2>
<p>In socialist systems, the government plays a larger role in economic decisions...</p>
<h2>8:20 - Conclusion</h2>
<p>Each system has tradeoffs. Understanding these helps us make informed decisions...</p>
</body>
</html>
Key features:
MM:SS - Title format (from video's chapter markers)