Content
```html
<div align="center">
# 🎯 Anything → NotebookLM Smart Processor
**One sentence to podcast, PPT, mind map, Quiz...**
[](https://opensource.org/licenses/MIT)
[](https://www.python.org/downloads/)
[](http://makeapullrequest.com)
[](https://github.com/joeseesun/anything-to-notebooklm/stargazers)
[](https://github.com/joeseesun/anything-to-notebooklm/network/members)
[](https://github.com/joeseesun/anything-to-notebooklm/issues)
[](https://github.com/joeseesun/anything-to-notebooklm/commits/main)
[Quick Start](#-quick-start) • [Supported Formats](#-supported-content-sources) • [Usage Examples](#-usage-examples) • [FAQ](#-faq)
</div>
---
## ✨ What is this?
A **Claude Code Skill** that lets you turn **anything** into **any format** using natural language.
```
You say: Generate a podcast from this WeChat article
AI: ✅ 8-minute podcast generated → podcast.mp3
You say: Make a mind map of this EPUB ebook
AI: ✅ Mind map generated → mindmap.json
You say: Turn this YouTube video into a PPT
AI: ✅ 25-page PPT generated → slides.pdf
```
**Principle**: Automatically get content from various sources → upload to [Google NotebookLM](https://notebooklm.google.com/) → AI generates the format you want
## 🚀 Supported Content Sources (15+ formats)
<table>
<tr>
<td width="50%">
### 📱 Social Media
- **WeChat Official Account** (bypass anti-crawler)
- **YouTube Videos** (automatic subtitle extraction)
### 🌐 Web
- **Any Webpage** (news, blogs, documents)
- **Search Keywords** (automatically summarize results)
### 📄 Office Documents
- **Word** (.docx)
- **PowerPoint** (.pptx)
- **Excel** (.xlsx)
</td>
<td width="50%">
### 📚 Ebooks and Documents
- **PDF** (supports scanned OCR)
- **EPUB** (ebook)
- **Markdown** (.md)
### 🖼️ Images and Audio
- **Images** (JPEG/PNG/GIF, automatic OCR)
- **Audio** (WAV/MP3, automatic transcription)
### 📊 Structured Data
- **CSV/JSON/XML**
- **ZIP Archive** (batch processing)
</td>
</tr>
</table>
**Technical Support**: [Microsoft markitdown](https://github.com/microsoft/markitdown)
## 🎨 What can be generated?
| Output Format | Use | Generation Time | Trigger Word Example |
|---------|------|---------|-----------|
| 🎙️ **Podcast** | Listen on the go | 2-5 minutes | "Generate a podcast", "Make it audio" |
| 📊 **PPT** | Team sharing | 1-3 minutes | "Make a PPT", "Generate slides" |
| 🗺️ **Mind Map** | Clarify structure | 1-2 minutes | "Draw a mind map", "Generate a mind map" |
| 📝 **Quiz** | Self-test mastery | 1-2 minutes | "Generate a Quiz", "Create questions" |
| 🎬 **Video** | Visualization | 3-8 minutes | "Make a video" |
| 📄 **Report** | In-depth analysis | 2-4 minutes | "Generate a report", "Write a summary" |
| 📈 **Infographic** | Data visualization | 2-3 minutes | "Make an infographic" |
| 📋 **Flashcards** | Memory consolidation | 1-2 minutes | "Make flashcards" |
**Completely natural language, no need to remember commands!**
## ⚡ Quick Start
### Prerequisites
- ✅ Python 3.9+
- ✅ Git (macOS/Linux comes with it)
**That's it!** Other dependencies are automatically installed with one click.
### Installation (3 steps)
```bash
# 1. Clone to Claude skills directory
cd ~/.claude/skills/
git clone https://github.com/joeseesun/anything-to-notebooklm
cd anything-to-notebooklm
# 2. One-click install all dependencies
./install.sh
# 3. Configure MCP as prompted, then restart Claude Code
```
### First Use
```bash
# NotebookLM authentication (only once)
notebooklm login
notebooklm list # Verify success
# Environment check (optional)
./check_env.py
```
## 💡 Usage Examples
### Scenario 1: Quick Learning - Article → Podcast
```
You: Generate a podcast from this article https://mp.weixin.qq.com/s/abc123
AI automatically executes:
✓ Grab WeChat article content
✓ Upload to NotebookLM
✓ Generate podcast (2-5 minutes)
✅ Result: /tmp/article_podcast.mp3 (8 minutes, 12.3 MB)
💡 Use: Listen to an in-depth article on the commute
```
### Scenario 2: Team Sharing - Ebook → PPT
```
You: Make this book into a PPT /Users/joe/Books/sapiens.epub
AI automatically executes:
✓ Extract ebook content (150,000 words)
✓ AI refines core viewpoints
✓ Generate professional PPT
✅ Result: /tmp/sapiens_slides.pdf (25 pages, 3.8 MB)
💡 Use: Directly used for book club sharing
```
### Scenario 3: Self-test Learning - Video → Quiz
```
You: Generate a Quiz from this YouTube video https://youtube.com/watch?v=abc
AI automatically executes:
✓ Extract video subtitles
✓ AI analyzes key knowledge points
✓ Automatically create questions
✅ Result: /tmp/video_quiz.md (15 questions, 10 multiple choice + 5 short answer)
💡 Use: Check learning effectiveness
```
### Scenario 4: Information Integration - Multi-source → Report
```
You: Make these contents into a report together:
- https://example.com/article1
- https://youtube.com/watch?v=xyz
- /Users/joe/research.pdf
AI automatically executes:
✓ Summarize 3 different sources
✓ AI integrates and analyzes
✓ Generate comprehensive report
✅ Result: /tmp/multi_source_report.md (7 chapters, 15.2 KB)
💡 Use: Comprehensive thematic research report
```
### Scenario 5: Document Digitization - Scan → Text
```
You: Make this scanned image into a document /Users/joe/scan.jpg
AI automatically executes:
✓ OCR recognizes the text in the image
✓ Extract as plain text
✓ Generate structured document
✅ Result: /tmp/scan_document.txt (95%+ recognition accuracy)
💡 Use: Digital archiving of scanned documents
```
## 🎯 Core Features
### 🧠 Intelligent Recognition
Automatically determine the input type, no need to manually specify
```
https://mp.weixin.qq.com/s/xxx → WeChat Official Account
https://youtube.com/watch?v=xxx → YouTube Video
/path/to/file.epub → EPUB Ebook
"Search 'AI Trends'" → Search Query
```
### 🚀 Fully Automatic Processing
From acquisition to generation, done in one go
```
Input → Get → Convert → Upload → Generate → Download
︿________Fully Automatic________︿
```
### 🌐 Multi-Source Integration
Supports mixing multiple content sources
```
Article + Video + PDF + Search Results → Comprehensive Report
```
### 🔒 Local Priority
Sensitive content is processed locally
```
WeChat Article → Local MCP Crawl → Local Conversion → NotebookLM
```
## 📦 Technical Architecture
```
┌─────────────────────────────────────┐
│ User Natural Language Input │
│ "Generate a podcast from this article https://..." │
└──────────────┬──────────────────────┘
│
▼
┌─────────────────────────────────────┐
│ Claude Code Skill │
│ • Intelligently identify content source type │
│ • Automatically call corresponding tools │
└──────────────┬──────────────────────┘
│
┌────────┴────────┐
│ │
▼ ▼
┌──────────┐ ┌─────────────┐
│ WeChat Official Account │ │ Other Formats │
│ MCP Crawl │ │ markitdown │
└─────┬────┘ └──────┬──────┘
│ │
└────────┬────────┘
│
▼
┌─────────────────────────────────────┐
│ NotebookLM API │
│ • Upload content source │
│ • AI generates target format │
└──────────────┬──────────────────────┘
│
▼
┌─────────────────────────────────────┐
│ Generated Files │
│ .mp3 / .pdf / .json / .md │
└─────────────────────────────────────┘
```
## 🔧 Advanced Usage
### Specify Existing Notebook
```
Add this article to my [AI Research] notebook https://example.com
```
### Batch Processing
```
Generate podcasts from these articles:
1. https://mp.weixin.qq.com/s/abc123
2. https://example.com/article2
3. /Users/joe/notes.md
```
### ZIP Batch Conversion
```
Make all the documents in this archive into podcasts /path/to/files.zip
```
Automatic decompression, identification, conversion, merging
## 🐛 Troubleshooting
### MCP Tool Not Found
```bash
# Test MCP Server
python ~/.claude/skills/anything-to-notebooklm/wexin-read-mcp/src/server.py
# Reinstall dependencies
cd ~/.claude/skills/anything-to-notebooklm/wexin-read-mcp
pip install -r requirements.txt
playwright install chromium
```
### NotebookLM Authentication Failed
```bash
notebooklm login # Re-login
notebooklm list # Verify
```
### Environment Check
```bash
./check_env.py # 13 comprehensive checks
./install.sh # Reinstall
```
## 🤝 Contribution
Welcome PRs, Issues, Suggestions!
## ❓ FAQ
<details>
<summary><b>Q: Which languages are supported?</b></summary>
A: NotebookLM supports multiple languages, with Chinese and English having the best results.
</details>
<details>
<summary><b>Q: Whose voice is the podcast?</b></summary>
A: Google AI speech synthesis. English is a dialogue between two AI hosts, while Chinese is a single person narration.
</details>
<details>
<summary><b>Q: Content length limit?</b></summary>
A:
- Minimum: Approximately 500 words
- Maximum: Approximately 500,000 words
- Recommended: 1000-10000 words for best results
</details>
<details>
<summary><b>Q: Can it be used commercially?</b></summary>
A:
- This Skill: MIT open source, free to use
- Generated content: Comply with NotebookLM Terms of Service
- Original content: Comply with original content copyright
- Suggestion: Only for personal learning and research
</details>
<details>
<summary><b>Q: Why is MCP needed?</b></summary>
A: WeChat Official Accounts have anti-crawler measures, MCP uses browser simulation to bypass. Other content sources (webpages, YouTube, PDF) do not require MCP.
</details>
## 📄 License
[MIT License](LICENSE)
## 🙏 Acknowledgements
- [Google NotebookLM](https://notebooklm.google.com/) - AI Content Generation
- [Microsoft markitdown](https://github.com/microsoft/markitdown) - File Conversion
- [wexin-read-mcp](https://github.com/Bwkyd/wexin-read-mcp) - WeChat Crawling
- [notebooklm-py](https://github.com/teng-lin/notebooklm-py) - NotebookLM CLI
## 📮 Contact
- **Issues**: [GitHub Issues](https://github.com/joeseesun/anything-to-notebooklm/issues)
- **Discussions**: [GitHub Discussions](https://github.com/joeseesun/anything-to-notebooklm/discussions)
---
<div align="center">
**If you find it useful, please give a ⭐ Star!**
Made with ❤️ by [Joe](https://github.com/joeseesun)
</div>
```
