notebooklm

Query NotebookLM notebooks for citation-backed answers from uploaded documents.

Updated Apr 25, 2026
One-click install
npx skills add https://github.com/quangkmhd/Vietstock_Agent --skill notebooklm-quangkmhd
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: notebooklm
Source: https://github.com/quangkmhd/Vietstock_Agent/tree/main/.agents/skills/notebooklm
Command: npx skills add https://github.com/quangkmhd/Vietstock_Agent --skill notebooklm-quangkmhd

SYSTEM DOCUMENTATION & REQUIREMENTS

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

What problem does it solve?

Query NotebookLM notebooks via Claude Code to deliver source-grounded, citation-backed answers drawn only from your uploaded documents.

Core Features & Use Cases

  • Flexible querying of NotebookLM notebooks through Claude Code, with automatic Google authentication and local notebook library management.
  • Stateless question workflow that opens a fresh browser session for each query, returns a grounded answer with citations, and prompts for follow-up to cover gaps.
  • Quick start includes adding notebooks via URL, running questions against the active notebook, and organizing work across multiple notebooks.

Quick Start

Open Claude Code, authenticate with Google NotebookLM through the browser window, add a NotebookLM URL to your library, then ask your first question.

Frequently Asked Questions about notebooklm

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

FAQPage Schema
How do I query NotebookLM through Claude Code to get grounded answers?

You can query NotebookLM through Claude Code by authenticating your Google account in a browser window, adding a notebook URL to your local library, and asking questions to receive source-grounded, citation-backed answers.

What is the best way to automate NotebookLM questions across multiple notebooks?

Automate NotebookLM queries across multiple notebooks by managing a local library and using a stateless workflow that opens a fresh browser session for each question to ensure isolated, reliable research.

Do I need to authenticate with Google every time I ask a question?

No, Google authentication is established initially, and the skill uses a stateless workflow that opens a fresh browser session for each question while maintaining your local notebook library for continuous querying.

Does this tool support browser automation for fetching citation-backed answers?

Yes, it uses browser automation via a Python-based workflow with a virtual environment and script wrappers, relying on the patchright dependency to automate browser actions for fetching citation-backed answers.

Why does the workflow open a fresh browser session for each NotebookLM query?

A fresh browser session is opened for each NotebookLM query to provide isolated, reliable research across multiple notebooks, ensuring stateless question execution prevents session conflicts and maintains data integrity.

What are the limitations of using a Python-based script wrapper for NotebookLM automation?

Limitations include dependency on the patchright library for browser automation and the requirement to maintain a local virtual environment, meaning all queried data and portable self-contained storage reside locally on your machine.