Skip to main content

Documentation Index

Fetch the complete documentation index at: https://mintlify.com/jacob-bd/notebooklm-mcp-cli/llms.txt

Use this file to discover all available pages before exploring further.

notebooklm-mcp-cli gives you full programmatic control over Google NotebookLM through two entry points: the nlm command-line interface for scripting and terminal workflows, and the notebooklm-mcp Model Context Protocol server that wires NotebookLM directly into AI assistants like Claude, Cursor, Gemini CLI, and GitHub Copilot. A single PyPI package installs both binaries, so you get the same underlying capabilities whether you are automating a pipeline in Bash or asking an AI agent to build a research notebook in natural language.

Two Ways to Use

The package ships two entry points that share the same feature set. nlm — Command-Line Interface Run nlm in any terminal for interactive use, shell scripts, or CI pipelines. Every feature available to the MCP server is also available as a CLI command.
nlm notebook list
nlm notebook create "Research Project"
nlm source add <notebook> --url "https://example.com/article"
nlm audio create <notebook> --confirm
nlm download audio <notebook> <artifact-id>
nlm share public <notebook>
notebooklm-mcp — MCP Server for AI Agents Connect any MCP-compatible AI assistant to NotebookLM with a single setup command. Once connected, you interact through natural language — the agent translates your instructions into the 35 available MCP tools.
nlm setup add claude-code
nlm setup add cursor
nlm setup add gemini
nlm setup add github-copilot

Installation

Install the package with uv, pip, or pipx. Python 3.11+ required.

Authentication

Authenticate with NotebookLM using Auto Mode or manual cookie extraction.

What Is NotebookLM?

NotebookLM is Google’s AI-powered research assistant. You upload sources — URLs, PDFs, Google Docs, YouTube videos, text snippets — and NotebookLM indexes them so you can query across all of them, generate audio overviews (podcasts), produce video explainers, create briefing docs, build slide decks, and more. Until now, all of that required the NotebookLM web UI. notebooklm-mcp-cli adds a programmatic layer on top. You can create notebooks, add sources, run queries, and trigger content generation entirely from your terminal or from an AI agent — and every query you send from the CLI or MCP automatically appears in your NotebookLM web UI chat history, so the two interfaces stay in sync.

Feature Overview

CapabilityCLI CommandMCP Tool
List notebooksnlm notebook listnotebook_list
Create notebooknlm notebook createnotebook_create
Add Sources (URL, Text, Drive, File)nlm source addsource_add
Query notebook (persists to web UI)nlm notebook querynotebook_query
Create Studio Content (Audio, Video, etc.)nlm studio createstudio_create
Revise slide decksnlm slides revisestudio_revise
Download artifactsnlm download <type>download_artifact
Web/Drive researchnlm research startresearch_start
Share notebooknlm share public/invitenotebook_share_*
Sync Drive sourcesnlm source syncsource_sync_drive
Batch operationsnlm batch query/create/deletebatch
Cross-notebook querynlm cross querycross_notebook_query
Pipelines (multi-step workflows)nlm pipeline run/listpipeline
Tag & smart selectnlm tag add/list/selecttag
Configure AI toolsnlm setup add/remove/list
Install AI Skillsnlm skill install/update
Diagnose issuesnlm doctor

Important Disclaimer

This project uses internal, undocumented Google APIs that may change without notice. It has been tested on free, Pro, and Google AI Ultra ($249/mo) tier accounts. NotebookLM Enterprise accounts may work but have not been formally tested. Use for personal and experimental purposes at your own risk.
Authentication is cookie-based — the tool extracts credentials directly from your browser session using the Chrome DevTools Protocol (CDP). There is no official Google API behind this integration.

Build docs developers (and LLMs) love