NotebookLM MCP Connector

Configure authenticated MCP access to Google NotebookLM for AI agents.

6|Updated Feb 17, 2026
One-click install
npx skills add https://github.com/ANONIMO432HZ/notebooklm-mcp-skill-setup --skill notebooklm-mcp-connector
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: NotebookLM MCP Connector
Source: https://github.com/ANONIMO432HZ/notebooklm-mcp-skill-setup/tree/main
Command: npx skills add https://github.com/ANONIMO432HZ/notebooklm-mcp-skill-setup --skill notebooklm-mcp-connector

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

It solves the problem of configuring authenticated access to your private NotebookLM content through an MCP server, so you can query and use NotebookLM data from your AI agent workflow without manual reconfiguration.

Core Features & Use Cases

  • MCP server setup guidance: Installs and wires the Python-based NotebookLM MCP server into your agent via mcp_config.json.
  • Authentication methods: Provides both an automatic cookie-extraction approach and a manual cookie/token extraction fallback to handle common 403/handshake issues.
  • Validation workflow: Includes a quick Python verification routine to confirm tokens are cached and notebooks can be listed.

Use case example: You want your agent (Claude/Cursor/Antigravity or any compatible MCP client) to list your notebooks and answer questions grounded in your NotebookLM collections.

Quick Start

Install the Skill and then run the authentication step (preferably notebooklm-mcp-auth) followed by configuring your mcp_config.json to launch notebooklm-mcp.

Frequently Asked Questions about NotebookLM MCP Connector

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

FAQPage Schema
How do I connect my AI agent to Google NotebookLM for grounded retrieval?

To connect your AI agent to Google NotebookLM, you install the notebooklm-mcp-server Python package and wire it into your agent via mcp_config.json, enabling authenticated queries and grounded retrieval from your notebooks.

Why does my NotebookLM MCP server return a 403 error during authentication?

A 403 error during NotebookLM MCP authentication usually means your session cookies are invalid or missing; you can resolve this by running the notebooklm-mcp-auth tool or manually generating a cookies.txt file to cache valid credentials.

Do I need Python to configure MCP access to NotebookLM?

Yes, you need Python 3.10 or higher to install the notebooklm-mcp-server package and run the authentication scripts required to configure MCP access to NotebookLM for your agent.

Can I use NotebookLM with Cursor or Claude through an MCP server?

Yes, you can use NotebookLM with agent environments like Cursor, Claude, or Antigravity by configuring the mcp_config.json file to launch the notebooklm-mcp server, allowing your agent to list notebooks and search your knowledge base.

What is the best way to verify that my NotebookLM MCP server is working?

The best way to verify your NotebookLM MCP server is working is to run the provided Python validation routine, which confirms that your authentication tokens are cached and successfully lists your available notebooks.

How does MCP cookie-based authentication work for NotebookLM?

MCP cookie-based authentication for NotebookLM works by extracting and caching your valid browser session tokens, either automatically via notebooklm-mcp-auth or manually through a cookies.txt file, to authorize server requests.