notebooklm

Query Google NotebookLM notebooks to generate citation-backed answers from uploaded documents.

14|5|Updated Jun 3, 2025

Search Tags:#research #source-grounded #Gemini #notebooklm #notebook #Google NotebookLM

Ready to run

npx skills add https://github.com/b-open-io/prompts --skill notebooklm-b-open-io

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: notebooklm
Source: https://github.com/b-open-io/prompts/tree/main/user/.claude/skills/notebooklm
Command: npx skills add https://github.com/b-open-io/prompts --skill notebooklm-b-open-io

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires patchright, python-dotenv, and includes scripts (resource) and references (resource) components.

What problem does it solve?

This skill queries Google NotebookLM notebooks via Claude Code to produce source-grounded, citation-backed answers from your uploaded documents.

Core Features & Use Cases

Fresh browser sessions: Each question starts a new session to avoid cross-task bleed
Smart discovery: Add notebooks by discovering their content via questions
Persistent library and auth: Manage notebooks, authentication, and active context

Quick Start

Check auth: python scripts/run.py auth_manager.py status
Add a notebook via URL: python scripts/run.py notebook_manager.py add --url "https://notebooklm.google.com/notebook/..." --name "My Notebook" --description "Contains..." --topics "topic1,topic2"
Ask a question: python scripts/run.py ask_question.py --question "Your question" --notebook-id notebook-id

Frequently Asked Questions about notebooklm

High-intent search queries and answers about installing and using this skill.

FAQPage Schema

How do I query research documents and get source-grounded answers?▼

Query NotebookLM notebooks through Claude Code to retrieve citation-backed answers directly from your uploaded documents. Each query starts a fresh browser session, ensuring answers are strictly grounded in your source material with full citations.

Can I add and manage multiple notebooks in NotebookLM from the command line?▼

Yes. Use the notebook manager to add notebooks by URL, assign names and topics, and maintain a persistent library. Authentication and active context are stored locally under ~/.claude/skills/notebooklm across sessions.

What's the best way to set up NotebookLM authentication?▼

Check authentication status with `python scripts/run.py auth_manager.py status`. The skill manages persistent auth credentials, allowing you to add notebooks and query them without re-authenticating for each question.

How do I discover and organize notebook content before querying it?▼

Add notebooks by asking discovery questions that reveal their content. Organize them with descriptive names, topics, and metadata. The skill learns notebook structure intelligently, enabling targeted queries against specific collections.

Does this work across multiple browser sessions without losing notebook data?▼

Yes. Persistent storage under ~/.claude/skills/notebooklm retains your notebook library and authentication state. Each query runs a fresh browser session to prevent cross-task interference while maintaining your configured notebooks.

Why use source-grounded queries instead of general search?▼

Source-grounded queries ensure every answer is backed by citations from your uploaded documents, eliminating hallucination. This is essential for research, compliance, and any work requiring verifiable, traceable information sources.