Click on "Install Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type @ followed by the MCP server name and your instructions, e.g., "@NotebookLM MCP Server create a new notebook about Tesla stock"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
NotebookLM의 기능을 MCP (Model Context Protocol)로 제공하는 서버입니다. AI 클라이언트(Claude Desktop, Cursor, Gemini CLI 등)에서 NotebookLM의 노트북 관리, 소스 추가 기능을 직접 사용할 수 있습니다.
노트북 목록 조회: list_notebooks
노트북 생성: create_notebook
웹 소스 추가: add_url_source
Related MCP server: notebooklm-mcp
이 프로젝트를 실행하려면 Python 3.10 이상이 필요합니다. Python 공식 홈페이지에서 설치해 주세요. 설치 시 "Add Python to PATH" 옵션을 반드시 체크해야 합니다.
(권장) 가상환경을 생성하고 활성화한 후 설치하는 것이 좋습니다.
python -m venv .venv
.\.venv\Scripts\Activate.ps1 # Windows PowerShell프로젝트 루트에서 다음 명령어를 실행합니다:
pip install .실행 오류가 발생하면 python -m pip install . 또는 uv pip install .을 시도해 보세요.
설치가 완료되면 다음 명령어로 MCP 서버를 실행할 수 있습니다:
notebooklm-mcp
# 또는
python -m notebooklm_mcpclaude_desktop_config.json 파일에 다음 설정을 추가하세요:
{
"mcpServers": {
"notebooklm": {
"command": "python",
"args": ["-m", "notebooklm_mcp"]
}
}
}더 자세한 실행 방법과 Claude/Cursor 연동 방법은 docs/usage.md 문서를 참고하세요.
Chrome 로그인 필요: 이 서버는 로컬 Chrome 브라우저의 쿠키를 사용하여 인증합니다. Chrome에서 Google 계정에 로그인된 상태여야 합니다.
쿠키 권한: 브라우저 종료 후 사용하는 것을 권장하며, 쿠키 접근 권한 오류 시 관리자 권한으로 실행하거나 전체 디스크 접근 권한이 필요할 수 있습니다.
List all NotebookLM notebooks owned by the user.
| Name | Required | Description | Default |
|---|---|---|---|
No parameters | |||
| Name | Required | Description |
|---|---|---|
| result | Yes |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
With no annotations, the description carries the full burden. It correctly implies a read-only operation ('list'), but does not disclose any behavioral traits such as pagination, ordering, or rate limits. The output schema exists but is not detailed here.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is a single sentence that front-loads the action and resource. Every word adds value, with no waste.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Given the tool has no parameters and an output schema exists (even if not shown), the description fully covers what the agent needs to know: it lists all notebooks owned by the user. No additional context is necessary for a simple list operation.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
There are zero parameters, so the description does not need to add parameter information. Baseline is 4 per instructions. The description is sufficient for a parameterless tool.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description uses a specific verb ('list') and resource ('NotebookLM notebooks') with scope ('owned by the user'), making the tool's purpose immediately clear. With no sibling tools, differentiation is not needed.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The description implies usage when a user wants to see their notebooks, but provides no explicit guidance on when or when not to use this tool, nor any alternatives. Since there are no siblings, the lack of exclusions is acceptable but not explicit.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
v0.1.0list_notebooksScored across 1 tool
With only one tool, there is no possibility of confusion between tools; the purpose is clear.
The single tool name 'list_notebooks' follows a clear verb_noun pattern, and consistency is not a concern with one tool.
A single tool for a NotebookLM server feels too thin; most users would expect additional operations beyond listing.
The server only offers listing, lacking essential operations like create, delete, or update, leaving the surface severely incomplete.
Unclaimed servers have limited discoverability.
If you are the server author, to access and configure the admin panel.
Google NotebookLM via natural language: create notebooks, add sources (PDF, URL, YouTube) and ask gr
Create, search and manage Knowtis collaborative notes from AI assistants.
Create, search, and update notes in an xNotepad AI notebook, with semantic search and AI Q&A.
AI access to your aNotepad online notes: read, search, write, and organize via 22 tools.
