Recommend an open-source project: qiaomu-anything-to-notebooklm. Someone built a hardcore tool using Claude, and after seeing it, I can only say: knowledge management enthusiasts should definitely bookmark this. You can throw in any content—WeChat public accounts, YouTube, podcasts, PDFs, Word, Excel, ebooks—it automatically organizes everything and finally generates podcasts, PPTs, mind maps. The best part is it operates entirely through natural language, no coding required. https://github.com/joeseesun/qiaomu-anything-to-notebooklm…

joeseesun/qiaomu-anything-to-notebooklm

Source: https://github.com/joeseesun/qiaomu-anything-to-notebooklm

🎯 Anything → NotebookLM

Multi-source content smart processor: Anything → Podcast / PPT / Mind Map / Quiz

License: MIT (https://opensource.org/licenses/MIT) Python 3.9+ (https://www.python.org/downloads/) PRs Welcome (http://makeapullrequest.com) GitHub stars (https://github.com/joeseesun/qiaomu-anything-to-notebooklm/stargazers) GitHub forks (https://github.com/joeseesun/qiaomu-anything-to-notebooklm/network/members) GitHub issues (https://github.com/joeseesun/qiaomu-anything-to-notebooklm/issues) GitHub last commit (https://github.com/joeseesun/qiaomu-anything-to-notebooklm/commits/main)

Quick Start • Supported Formats • Usage Examples • Paywall Bypass • FAQ

✨ What is this?

A Claude Code Skill that turns anything into any format using natural language.

You: Turn this WeChat article into a podcast
AI: ✅ 8-minute podcast generated → podcast.mp3

You: Turn this paywalled article into a mind map
AI: ✅ Automatically bypassed paywall → Mind map generated

You: Turn this podcast (Xiaoyuzhou) into a PPT
AI: ✅ Auto-transcribed audio → 25-slide PPT generated

Core capability: Multi-source content acquisition (including paywall bypass) → Upload to Google NotebookLM (https://notebooklm.google.com/) → AI generates target format

🚀 Supported Content Sources (15+)

📱 Social & Media

• WeChat public accounts (MCP browser simulation)
• X/Twitter (tweets + long threads)
• YouTube videos (auto-extract subtitles)
• Podcasts (Xiaoyuzhou / Ximalaya / Bilibili)

🌐 Web Pages (including paywall bypass)

• 300+ paywalled sites (NYT/WSJ/FT/Economist…)
• Any public web page (news, blogs, docs)
• Search keywords (auto-summarize results)

📚 eBooks & Documents

• PDF (supports OCR for scanned documents)
• EPUB ebooks
• Markdown (.md)
• Plain text (.txt)

📄 Office Documents

• Word (.docx)
• PowerPoint (.pptx)
• Excel (.xlsx)

🖼️ Others

• Images (JPEG/PNG, auto OCR)
• Audio (WAV/MP3, auto transcription)
• ZIP archives (batch processing)

🛡️ Paywall Bypass

Core feature: Automatically detects and bypasses paywalls on 300+ paid news sites.

Bypass Strategy (6-level cascade)

Level 1: Proxy service (r.jina.ai / defuddle.md)
↓ Fail
Level 2: Site-specific Bot UA (Googlebot ~50 sites / Bingbot ~4 sites)
↓ Fail
Level 3: Generic bypass (UA spoofing + X-Forwarded-For + Referer spoofing + AMP + EU IP)
↓ Fail
Level 4: archive.today snapshot (auto CAPTCHA detection)
↓ Fail
Level 5: Google Cache
↓ Fail
Level 6: agent-fetch local tool

Supported Paid Sites (partial)

Bypass Techniques (learned from Bypass Paywalls Clean (https://gitflic.ru/project/magnolia1234/bypass-paywalls-chrome-clean))

🎨 What can be generated?

⚡ Quick Start

Prerequisites

• ✅ Python 3.9+
• ✅ Git (built-in on macOS/Linux)

That’s all! Other dependencies are installed automatically with one command.

Installation (3 steps)

# 1. Clone into Claude skills directory
cd ~/.claude/skills/
git clone https://github.com/joeseesun/qiaomu-anything-to-notebooklm
cd qiaomu-anything-to-notebooklm

# 2. One-click install all dependencies
./install.sh

# 3. Configure MCP as prompted, then restart Claude Code

First Use

# NotebookLM authentication (only once)
notebooklm login
notebooklm list   # Verify success

# Environment check (optional)
./check_env.py

Podcast Transcription Configuration (Optional)

To use Xiaoyuzhou / Ximalaya / Bilibili transcription, configure Get Notes API:

export GETNOTE_API_KEY="your_api_key"
export GETNOTE_CLIENT_ID="your_client_id"

💡 Usage Examples

Scenario 1: Paywalled Article → Podcast

You: Turn this The Information article into a podcast
https://www.theinformation.com/articles/...

AI auto-executes:
 ✓ Detect paywall → Bypass with Googlebot UA
 ✓ Get full article content
 ✓ Upload to NotebookLM
 ✓ Generate podcast
 ✅ Result: /tmp/article_podcast.mp3

Scenario 2: Podcast (Xiaoyuzhou) → PPT

You: Turn this Xiaoyuzhou podcast into a PPT
https://xiaoyuzhoufm.com/episode/...

AI auto-executes:
 ✓ Transcription via Get Notes API (2-5 minutes)
 ✓ Upload transcript to NotebookLM
 ✓ Generate PPT
 ✅ Result: /tmp/podcast_slides.pdf (25 slides)

Scenario 3: eBook → Deep Analysis

You: Deeply analyze this book /Users/joe/Books/sapiens.epub

AI auto-executes:
 ✓ Extract full EPUB text
 ✓ Upload to NotebookLM
 ✓ Generate 12 questions (3 progressive rounds: overview → deep dive → synthesis)
 ✓ Ask round by round, later rounds benefit from previous conversation context
 ✓ Output structured JSON
 ✅ Result: /tmp/sapiens_analysis.json (12 Q&As including core points, argument breakdown, contradictions, cognitive shifts)

Scenario 4: X/Twitter Thread → Mind Map

You: Turn this tweet thread into a mind map
https://x.com/user/status/123...

AI auto-executes:
 ✓ Proxy cascade to fetch tweet content (including full thread)
 ✓ Upload to NotebookLM
 ✓ Generate mind map
 ✅ Result: /tmp/tweet_mindmap.json

Scenario 5: WeChat Article → Feishu Document (Deep Analysis)

You: Deeply analyze this WeChat article and write to Feishu
https://mp.weixin.qq.com/s/abc123

AI auto-executes:
 ✓ MCP browser simulation to scrape WeChat article
 ✓ Upload to NotebookLM
 ✓ Generate 10 questions and recursively ask
 ✓ Format as Feishu Markdown
 ✓ Auto-create Feishu document
 ✅ Result: Feishu document created (with full Q&A)

🎯 Core Features

🧠 Smart Recognition

Automatically determines input type, no manual specification needed

https://mp.weixin.qq.com/s/xxx       → WeChat public account
https://xiaoyuzhoufm.com/episode/xxx → Xiaoyuzhou podcast
https://x.com/user/status/xxx       → X/Twitter
https://youtube.com/watch?v=xxx     → YouTube video
/path/to/file.epub                  → EPUB ebook
"Search 'AI trends'"                → Search query

🛡️ Auto Paywall Bypass

No manual handling needed, auto-detect and bypass

Detect paywall → Choose best strategy → Get full content
〈________Fully automatic________〈

🚀 Fully Automated Processing

From acquisition to generation, all in one go

Input → Acquire → Convert → Upload → Generate → Download
〈___________Fully automatic___________〈

🌐 Multi-Source Integration

Supports mixing multiple content sources

Paywalled article + YouTube video + EPUB + Podcast → Comprehensive report

📦 Technical Architecture

┌──────────────────────────────────────────┐
│ User natural language input              │
│ "Turn this paywalled article into a      │
│  podcast https://..."                    │
└──────────────────┬───────────────────────┘
                   │
                   ▼
┌──────────────────────────────────────────┐
│ Claude Code Skill                        │
│ • Smart content source type recognition  │
│ • Auto-call corresponding tools          │
└──────────┬───────────────────────────────┘
           │
    ┌──────┴───────┐
    │              │
    ▼              ▼
┌──────────┐ ┌──────────────┐ ┌──────────┐ ┌──────────┐
│WeChat MCP│ │Paywall Bypass│ │Podcast   │ │markitdown│
│Browser   │ │6-level       │ │Transcrip │ │File      │
│Simulation│ │cascade       │ │Get Notes │ │Conversion│
└─────┬────┘ └──────┬───────┘ │API       │ └─────┬────┘
      │             │         └─────┬────┘       │
      └──────────────┴─────────────────┴──────────┘
                                │
                                ▼
                    ┌────────────────────────┐
                    │ NotebookLM API         │
                    │ • Upload content source │
                    │ • AI generates target  │
                    │   format               │
                    └───────────┬────────────┘
                                │
                                ▼
                    ┌────────────────────────┐
                    │ Generated files        │
                    │ .mp3 / .pdf / .json    │
                    └────────────────────────┘

📂 Project Structure

qiaomu-anything-to-notebooklm/
├── SKILL.md                 # Skill definition file
├── README.md                # This file
├── main.py                  # Main entry: CLI smart processor
├── install.sh               # One-click install script
├── check_env.py             # 13-item environment check
├── package.sh               # Package and share script
├── requirements.txt         # Python dependencies
├── LICENSE                  # MIT
├── scripts/
│   ├── fetch_url.sh         # URL fetching + paywall bypass (6-level cascade)
│   └── get_podcast_transcript.py  # Podcast/video transcription (Get Notes API)
├── wexin-read-mcp/          # WeChat public account MCP server
│   └── src/
│       ├── server.py        # MCP entry
│       ├── scraper.py       # Playwright browser simulation
│       └── parser.py        # HTML parsing
└── feishu-read-mcp/         # Feishu document MCP server
    └── src/
        ├── server.py        # MCP entry
        ├── scraper.py       # Feishu document scraping
        ├── parser.py        # HTML → Markdown
        └── image_handler.py # Image processing

🔧 Advanced Usage

Deep Analysis Mode

python main.py https://example.com/article --deep-analysis
# Auto-generates 12 questions (3 progressive rounds: overview → deep dive → synthesis), asks round by round, outputs structured JSON

Three-round progressive strategy:

NotebookLM maintains context within the same session, so later round questions automatically benefit from earlier answers, forming a true “progressive” deep analysis.

Feishu Document Output

python main.py ./book.epub --deep-analysis --to-feishu
# After deep analysis, automatically create a Feishu document

Batch Processing

Turn all these articles into podcasts:
1. https://mp.weixin.qq.com/s/abc123
2. https://www.wsj.com/articles/...
3. /Users/joe/notes.md

🐛 Troubleshooting

MCP Tool Not Found

python ~/.claude/skills/qiaomu-anything-to-notebooklm/wexin-read-mcp/src/server.py
cd ~/.claude/skills/qiaomu-anything-to-notebooklm/wexin-read-mcp
pip install -r requirements.txt
playwright install chromium

NotebookLM Authentication Failed

notebooklm login   # Re-login
notebooklm list    # Verify

Paywall Bypass Failed

Some hard paywalled sites (e.g., The Information) do not send content server-side; an archive.today snapshot is needed. The script will auto-detect and prompt:

⚠️ archive.ph needs human verification.
Browser opened automatically. Please complete the verification and retry.

Environment Check

./check_env.py   # 13 comprehensive checks
./install.sh     # Reinstall

❓ FAQ

Q: Which languages are supported? A: NotebookLM supports multiple languages; Chinese and English work best.

Q: Who voices the podcast? A: Google AI voice synthesis. English has two AI hosts in conversation; Chinese is a single narrator.

Q: Is paywall bypass legal? A: This tool is for personal study and research only. The technique is based on search engine whitelists (Googlebot/Bingbot) and does not crack any encryption. We recommend supporting quality news media by purchasing subscriptions.

Q: Content length limits? A:

• Shortest: ~500 words
• Longest: ~500,000 words
• Recommended: 1000-10,000 words for best results

Q: Why is MCP needed? A: WeChat public accounts have anti-scraping measures; MCP uses Playwright browser simulation to bypass. Other content sources (web pages, YouTube, PDF) do not require MCP.

Q: Which platforms are supported for podcast transcription? A: Through Get Notes API, supports Xiaoyuzhou, Ximalaya, Bilibili videos. YouTube is handled directly by NotebookLM.

🙏 Acknowledgments

• Google NotebookLM (https://notebooklm.google.com/) - AI content generation
• Microsoft markitdown (https://github.com/microsoft/markitdown) - File conversion
• Bypass Paywalls Clean (https://gitflic.ru/project/magnolia1234/bypass-paywalls-chrome-clean) - Paywall bypass strategy reference
• wexin-read-mcp (https://github.com/Bwkyd/wexin-read-mcp) - WeChat scraping
• notebooklm-py (https://github.com/teng-lin/notebooklm-py) - NotebookLM CLI

📄 License

MIT License - For personal study and research only

If you find it useful, please give it a ⭐ Star!

Made with ❤️ by Joe (https://github.com/joeseesun) · Twitter @vista8 (https://x.com/vista8) · WeChat public account「向阳乔木推荐看」