Claude Code + NotebookLM

Connect two of the most powerful AI tools available. Run end-to-end research workflows — ingest sources, chat, generate artifacts, and download outputs — directly from your terminal.

tools claude-code Last updated: 2026-03-05

What makes this integration work: notebooklm-py

The open-source notebooklm-py project is the open-source tool that powers this setup. It provides the Python script for connecting to NotebookLM and the Claude Skill for proper use through the terminal.

🐍

Python tooling

Connects to NotebookLM from any terminal environment.

Claude Code skill

Installs as a skill so Claude can invoke it with natural language.

🔧

Full CLI

Notebook creation, source ingestion, analysis, generation, and downloads.

Features available through notebooklm-py

notebooklm-py exposes the full NotebookLM suite, including notebook creation, analysis, pitch deck generation, and more.

Studio outputs

Audio Overview
Video Overview
Mind Map
Reports
Flashcards
Quiz
Infographic
Slide Deck
Data Table

Installation

Just open your terminal and run the short code snippet below. The installation only takes a few seconds, and a browser window will pop up asking you to log into NotebookLM the first time.

bash
# Basic installation
pip install notebooklm-py

# With browser login support (required for first-time setup)
pip install "notebooklm-py[browser]"
playwright install chromium

# 1. Authenticate (opens browser)
notebooklm login

# Install via CLI or ask Claude Code to do it
notebooklm skill install

Using the skill with Claude Code

Claude Code links to NotebookLM through the notebooklm skill. You just describe what you want in natural language, and it automatically triggers the skill to carry it out.

2) Create a notebook and add sources

bash
notebooklm create "My Research"
notebooklm use <notebook_id>
notebooklm source add "https://en.wikipedia.org/wiki/Artificial_intelligence"
notebooklm source add "./paper.pdf"

3) Chat with your sources

bash
notebooklm ask "What are the key themes?"

4) Generate content

bash
notebooklm generate audio "make it engaging" --wait
notebooklm generate video --style whiteboard --wait
notebooklm generate quiz --difficulty hard
notebooklm generate flashcards --quantity more
notebooklm generate slide-deck
notebooklm generate infographic --orientation portrait
notebooklm generate mind-map
notebooklm generate data-table "compare key concepts"

5) Download artifacts

bash
notebooklm download audio ./podcast.mp3
notebooklm download video ./overview.mp4
notebooklm download quiz --format markdown ./quiz.md
notebooklm download flashcards --format json ./cards.json
notebooklm download slide-deck ./slides.pdf
notebooklm download mind-map ./mindmap.json
notebooklm download data-table ./data.csv

Best-fit use cases

This pairing works especially well for:

▶️

YouTube research

Ingest a video URL and let NotebookLM extract structured insights automatically.

📊

Slide decks & infographics

Turn source material into presentation-ready deliverables in one command.

🗂️

Flashcards & quizzes

Generate study materials from any document, PDF, or URL.

Copy/paste templates for Claude Code terminal

Replace [INSERT YOUTUBE URL HERE] before running.

Option 1: Executive Summary (Slide Deck)

Use the notebooklm skill to create a new notebook called 'Video Research'. Add this YouTube video as a source: [INSERT YOUTUBE URL HERE]. Once it's processed, generate a comprehensive slide deck that outlines the key takeaways and download the PDF to my current folder.

Option 2: Deep Study (Flashcards + Quiz)

I need to learn the material in this video: [INSERT YOUTUBE URL HERE]. Use the notebooklm skill to add it to a new notebook. Generate a quiz with 'hard' difficulty and a set of flashcards for the core concepts. Download both files in markdown or JSON format.

Option 3: Visual Breakdown (Infographic + Mind Map)

Connect to NotebookLM and ingest this video: [INSERT YOUTUBE URL HERE]. Use the tool to generate a mind map of the video's structure and a portrait-oriented infographic summarizing the data. Save the artifacts to my workspace.

Pro tip

💡

If you have already run notebooklm login in your terminal, Claude can access your existing notebooks. To use an existing notebook instead of creating a new one, prompt Claude like this:

"Use my existing 'Project X' notebook to add this video..."

References