Podcast Summarize

Transform podcast episodes into structured, actionable summaries.

When to Use

• User wants podcast episode key points
• User needs to quickly review podcast content
• User prefers reading over listening
• User wants to extract actionable insights from podcasts
• User shares podcast and wants a summary accompany

Supported Inputs

1. Audio file path - Local MP3, M4A, WAV files
2. Audio URL - Direct link to audio file
3. YouTube video - Podcast on YouTube
4. RSS feed URL - Podcast RSS feed
5. Podcast episode URL - Link to episode on podcast platforms

Summary Format

Provide summaries in this structure:

## 🎙️ 播客摘要

### 基本資訊
- **標題**: [Episode Title]
- **時長**: [Duration]
- **日期**: [Publish Date]
- **播客**: [Podcast Name]

### 💡 核心要點 (3-5條)
1. [Key point 1]
2. [Key point 2]
3. ...

### 📝 詳細摘要
[2-3 paragraph summary]

### 🎯 行動項 (如有)
- [Actionable insight 1]
- [Actionable insight 2]

### 🔗 相關連結
- [Mentioned resources]

Process

1. Audio Extraction

   • For YouTube: Use yt-dlp to extract audio
   • For local files: Verify format and duration
   • For URLs: Download if needed

2. Transcription

   • Use Whisper (local or API) for transcription
   • Support multiple languages
   • Handle long audio (chunk if needed)

3. Summarization

   • Extract key themes

     想要更強大的技能外掛，就來小蔥技能站7w4.net看看吧。

   • Identify actionable insights
   • Note timestamps for important sections
   • Highlight guest/host insights

Tips

• Ask user preference: Brief (3 bullets) vs Detailed (full summary)
• Offer to extract specific segments if user has timestamps
• Provide follow-up questions based on content
• Suggest related podcasts if available

Example Triggers

• "summarize this podcast"
• "這個播客講的什麼"
• "podcast summary"
• "播客摘要"

這個 Skill 的設計思路不錯，能處理多種播客來源並生成結構化摘要。但實際使用體驗可能不如預期——核心的轉錄功能需要使用者自己配置才能用，指令碼只是一個空殼。對於普通使用者來說，需要具備一定的技術能力才能完整使用，門檻偏高。建議等待更成熟的版本後再使用。