notebooklm-skill

Overview

NotebookLM-skill bridges Google's NotebookLM research capabilities with Claude's content generation, creating a complete workflow from research discovery to content publishing. It processes various source types (URLs, PDFs, YouTube videos, trending topics) through NotebookLM notebooks, extracts structured insights, and hands them to Claude for polished output in multiple formats. The server offers both standalone MCP functionality and integration as a Claude Code skill, with a focus on automation through pipeline workflows.

Try asking AI

After installing, here are 7 things you can ask your AI assistant:

• nlm_create_notebook

  Create a notebook with URL/text sources

  Try:Create a notebook about AI ethics with these sourcesTry:Make a new notebook with this research paper
• nlm_list

  List all notebooks

  Try:Show me all my notebooksTry:List available notebooks
• nlm_delete

  Delete a notebook

  Try:Remove the old AI research notebookTry:Delete this notebook
• nlm_add_source

  Add a source to existing notebook

  Try:Add this PDF to my climate change notebookTry:Include this article in my research
• nlm_ask

  Ask a research question (returns answer + citations)

  Try:What are the key findings about this topic?Try:Ask about the main arguments in this research
• nlm_summarize

  Get notebook summary

  Try:Summarize my research notebookTry:Give me a quick overview of this notebook
• nlm_generate

  Generate artifact (audio, video, slides, etc.)

  Try:Create a podcast from this researchTry:Generate slides for this notebookTry:Make a video summary
• nlm_download

  Download generated artifact

  Try:Download the audio fileTry:Get the slides PDF
• nlm_list_sources

  List sources in notebook

  Try:Show me all sources in this notebookTry:What articles are included?
• nlm_list_artifacts

  List generated artifacts

  Try:What artifacts have been generated?Try:List all created files
• nlm_research

  Deep web research

  Try:Run deep research on this topicTry:Find more information about this subject
• nlm_research_pipeline

  Research to content pipeline

  Try:Create an article from this researchTry:Generate social posts from this notebook

Comparable tools

Installation

uvx --from notebooklm-skill notebooklm-mcp  # Start MCP server

pip install notebooklm-skill
notebooklm-mcp  # Start MCP server

git clone https://github.com/claude-world/notebooklm-skill.git
cd notebooklm-skill && pip install .

{
  "mcpServers": {
    "notebooklm": {
      "command": "uvx",
      "args": ["--from", "notebooklm-skill", "notebooklm-mcp"]
    }
  }
}

# One-time login (opens Chromium, sign in with Google)
uvx notebooklm login
# or: python3 -m notebooklm login

FAQ

Does this require Google API keys or OAuth setup?

No. It uses browser-based Google login with session storage, requiring no API keys or OAuth Client ID setup.

What output formats are supported?

It generates multiple artifacts including audio (M4A), video (MP4), slides (PDF/PPTX), reports (Markdown), quizzes, flashcards, mind maps, infographics, data tables, and study guides.

Can it work with other MCP clients besides Claude?

Yes. It works as a standalone MCP server compatible with any MCP client like Cursor, Gemini CLI, etc.

Compare notebooklm-skill with