notebooklm-skill

Query NotebookLM notebooks to synthesize grounded answers from uploaded documents.

Updated Feb 12, 2026

Search Tags:#research #notebooklm #browser-automation #grounded #cross-source #document-library

Ready to run

npx skills add https://github.com/arendon1/agent-skills --skill notebooklm-skill

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: notebooklm-skill
Source: https://github.com/arendon1/agent-skills/tree/main/notebooklm-skill
Command: npx skills add https://github.com/arendon1/agent-skills --skill notebooklm-skill

SYSTEM DOCUMENTATION & REQUIREMENTS

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

What problem does it solve?

Helps AI agents perform deep, multi-source research by querying NotebookLM and grounding results in uploaded documents.

Core Features & Use Cases

Grounded cross-document synthesis across PDFs, Docs, and web sources using NotebookLM.
Discover, ingest, and organize notebooks with metadata and citations.
Generate grounded artifacts such as summaries, mind maps, quizzes, and grounded reports for study aids or decision support.

Quick Start

To start, run a query against a NotebookLM notebook using the unified bridge to retrieve a grounded answer.

Frequently Asked Questions about notebooklm-skill

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

FAQPage Schema

How do I automate cross-document research using NotebookLM?▼

Automate cross-document research by using a script-based bridge to query NotebookLM, grounding synthesized answers in your uploaded PDFs, docs, and web sources. This enables agents to retrieve cited multi-source summaries programmatically.

What is grounded cross-source synthesis and how does it work with uploaded documents?▼

Grounded cross-source synthesis queries your NotebookLM document library to generate answers strictly based on uploaded PDFs, docs, and web sources. It ensures agent responses are backed by citations from your local data store.

Can I generate summaries and mind maps from a document library through browser automation?▼

Yes, you can generate summaries and mind maps from your document library through browser automation. The Skill uses a modular script architecture to drive NotebookLM, producing grounded artifacts like quizzes and reports.

Do I need patchright to run NotebookLM browser automation scripts?▼

Yes, patchright is required as a dependency to run the browser automation scripts. The Skill relies on it alongside notebooklm-py to execute deterministic queries and manage notebook state locally.

What is the best way to organize and query multiple notebooks for AI agents?▼

The best way to organize and query multiple notebooks is using the local data store for libraries and state. The modular run wrapper executes deterministic scripts to discover, ingest, and query notebooks with metadata and citations.

Why does grounded document synthesis return inconsistent results during automated queries?▼

Inconsistent grounded document synthesis during automated queries often stems from non-deterministic execution. The Skill solves this by using a modular script architecture with a run wrapper to ensure deterministic execution and stable state management.