notebooklm-mcp-server

notebooklm-mcp-server

Enables AI agents to interact with Google NotebookLM for grounded, hallucination-free answers through notebook management, source management, research, and generation tools.

Category
访问服务器

README

<p align="center"> <img src="./notebooklm_mcp_logo.png" width="200" alt="Notebook-mcp-server Logo"> </p>

<h1 align="center">NotebookLM MCP Server</h1>

<p align="center"> <b>Let your AI agents chat directly with Google NotebookLM for zero-hallucination answers.</b> </p>

<p align="center"> <b>English</b> • <a href="README.es.md">Español</a> • <a href="README.fr.md">Français</a> • <a href="README.pt.md">Português</a> • <a href="README.de.md">Deutsch</a> </p>

<p align="center"> <a href="https://www.typescriptlang.org/"><img src="https://img.shields.io/badge/TypeScript-3178C6?style=for-the-badge&logo=typescript&logoColor=white" alt="TypeScript"></a> <a href="https://modelcontextprotocol.io"><img src="https://img.shields.io/badge/MCP-Model%20Context%20Protocol-orange?style=for-the-badge" alt="MCP"></a> <a href="https://www.npmjs.com/package/notebooklm-mcp-server"><img src="https://img.shields.io/badge/NPM-CB3837?style=for-the-badge&logo=npm&logoColor=white" alt="NPM"></a> </p>

<p align="center"> <img src="https://img.shields.io/badge/Windows-0078D4?style=for-the-badge&logo=windows&logoColor=white" alt="Windows"> <img src="https://img.shields.io/badge/macOS-000000?style=for-the-badge&logo=apple&logoColor=white" alt="macOS"> <img src="https://img.shields.io/badge/Linux-FCC624?style=for-the-badge&logo=linux&logoColor=black" alt="Linux"> </p>

<p align="center"> <a href="https://anthropic.com"><img src="https://img.shields.io/badge/Claude%20Code-Skill-blueviolet?style=for-the-badge" alt="Claude Code"></a> <a href="https://geminicli.com/"><img src="https://img.shields.io/badge/Gemini%20CLI-Skill-blueviolet?style=for-the-badge" alt="Gemini CLI"></a> <img src="https://img.shields.io/badge/Cursor-000000?style=for-the-badge&logo=cursor&logoColor=white" alt="Cursor"> <img src="https://img.shields.io/badge/Windsurf-00AEEF?style=for-the-badge" alt="Windsurf"> <img src="https://img.shields.io/badge/Cline-FF5733?style=for-the-badge" alt="Cline"> </p>

<p align="center"> <a href="#installation">Installation</a> • <a href="#authentication">Authentication</a> • <a href="#quick-start-claude-desktop">Quick Start</a> • <a href="#claude-code-skill">Claude Code</a> • <a href="#documentation">Documentation</a> • <a href="#development">Development</a> </p>

The Solution

The NotebookLM MCP Server brings the power of Google's NotebookLM directly into your AI-augmented workflow. Built natively in TypeScript using the Model Context Protocol, it allows agents to read, search, and manage your notebooks as if they were local files.


🚀 Installation

1. Global Installation (Recommended)

You can install the server directly from NPM:

npm install -g notebooklm-mcp-server

[!NOTE] Auto-update: The server automatically checks for new versions at startup. If an update exists, it will install itself and ask you to restart to ensure you always have the latest Google fixes.

2. Direct usage with NPX (Zero-Config)

If you don't want to install it globally, you can run it directly:

npx notebooklm-mcp-server auth   # To log in
npx notebooklm-mcp-server start  # To run the server

🔑 Authentication

Before using the server, you must link it to your Google Account. This version uses a secure, persistent browser session:

  1. Run the authentication command:
    npx notebooklm-mcp-server auth
    
  2. A browser window will open. Log in with your Google account.
  3. Close the browser once you see your notebooks. Your session is now securely saved locally.

[!TIP] Session Expired? If your agent receives authentication errors, simply ask it to run the command npx notebooklm-mcp-server refresh_auth. It will automatically open the browser for you to renew the session without leaving your chat.


⚡ Quick Start

🤖 Claude Desktop

Add the following to your claude_desktop_config.json:

{
  "mcpServers": {
    "notebooklm": {
      "command": "npx",
      "args": ["-y", "notebooklm-mcp-server", "start"]
    }
  }
}

💻 Visual Studio Code

Since VS Code does not support MCP natively yet, you must use an extension:

Option A: Using Cline (Recommended)

  1. Open Cline Settings in VS Code.
  2. Scroll to the MCP Servers section.
  3. Click Add New MCP Server.
  4. Use the following configuration:
    • Name: notebooklm
    • Command: npx -y notebooklm-mcp-server start

Option B: Using MCP Client

  1. Install the extension from the Marketplace.
  2. Open your VS Code settings.json.
  3. Add the server under mcp.servers:
    "mcp.servers": {
      "notebooklm": {
        "command": "npx",
        "args": ["-y", "notebooklm-mcp-server", "start"]
      }
    }
    

🌌 Antigravity

Antigravity supports MCP natively. You can add the server by editing your global configuration file:

  1. Locate your mcp.json:

    • Windows: %APPDATA%\antigravity\mcp.json
    • macOS: ~/Library/Application Support/antigravity/mcp.json
    • Linux: ~/.config/antigravity/mcp.json
  2. Add the server to the mcpServers object:

{
  "mcpServers": {
    "notebooklm": {
      "command": "npx",
      "args": ["-y", "notebooklm-mcp-server", "start"]
    }
  }
}
  1. Restart Antigravity: The new tools will appear in your sidebar instantly.

💎 Gemini CLI

Run the following command in your terminal to add the notebooklm skill:

gemini mcp add notebooklm --scope user -- npx -y notebooklm-mcp-server start

🤖 Claude Code Skill

Add it instantly to Claude Code:

claude skill add notebooklm -- "npx -y notebooklm-mcp-server start"

📖 Documentation

The following tools are available through this MCP server:

📒 Notebook Management

Tool Description
notebook_list Lists all notebooks in your account.
notebook_create Creates a new notebook with a title.
notebook_rename Renames an existing notebook.
notebook_delete Deletes a notebook (Warning: Destructive).

🖇️ Source Management

Tool Description
notebook_add_url Adds a website or YouTube video as a source.
notebook_add_text Adds custom text content as a source.
notebook_add_local_file Uploads a local PDF, Markdown or Text file.
notebook_add_drive Adds a file from Google Drive (Docs, Slides, etc).
source_delete Removes a source from a notebook.
source_sync Syncs a Drive source to get the latest version.

🔍 Research & Query

Tool Description
notebook_query Asks a grounded question to a specific notebook.
research_start Starts a web/drive research task.
research_poll Polls for research status and results.
research_import Imports research results as permanent sources.

🎨 Studio & Generation

Tool Description
audio_overview_create Generates an Audio Overview (podcast).
studio_poll Checks status of generated audio/video artifacts.
mind_map_generate Generates a Mind Map JSON from sources.

⚙️ System

Tool Description
refresh_auth Interactive: Opens a browser to renew your Google session. Use this if tools start failing.

🛠️ Development

To contribute or build from source:

git clone https://github.com/moodRobotics/notebook-mcp-server.git
npm install
npm run build

🌐 Localization

This project is available in multiple languages:

If you update the English README.md, please ensure the translations are updated accordingly to keep the documentation synchronized.

📄 License

MIT License. Developed with ❤️ by moodRobotics.

推荐服务器

Baidu Map

Baidu Map

百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。

官方
精选
JavaScript
Playwright MCP Server

Playwright MCP Server

一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。

官方
精选
TypeScript
Magic Component Platform (MCP)

Magic Component Platform (MCP)

一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。

官方
精选
本地
TypeScript
Audiense Insights MCP Server

Audiense Insights MCP Server

通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。

官方
精选
本地
TypeScript
VeyraX

VeyraX

一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。

官方
精选
本地
graphlit-mcp-server

graphlit-mcp-server

模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。

官方
精选
TypeScript
Kagi MCP Server

Kagi MCP Server

一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。

官方
精选
Python
e2b-mcp-server

e2b-mcp-server

使用 MCP 通过 e2b 运行代码。

官方
精选
Neon MCP Server

Neon MCP Server

用于与 Neon 管理 API 和数据库交互的 MCP 服务器

官方
精选
Exa MCP Server

Exa MCP Server

模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。

官方
精选