This is an early release preview. You may encounter bugs.
Gemini Notebook MCP CLI logo
Unclaimed

Tool coding

Gemini Notebook MCP CLI

Programmatic access to Gemini Notebook - via command-line interface (CLI), Model Context Protocol (MCP) server, and AI agent skills.

A 87/100 GitHub score ? This grade is derived from GitHub signals, not user votes. Open for the full breakdown.
Discussion

01 / Summary

About Gemini Notebook MCP CLI.

Gemini Notebook MCP CLI gives scripts and AI agents access to Gemini Notebook, the Google research-notebook service formerly called NotebookLM. One package ships two front ends over the same capabilities: the nlm command-line tool for terminal use and automation, and an MCP server that exposes 53 tools to AI assistants, so a request such as creating a notebook on a topic and generating a podcast from it runs from natural language.

It works by calling the service's internal, undocumented APIs with cookies taken from a signed-in browser session. Those APIs can change without notice and may not align with Google's terms of service, and the project positions itself for personal and experimental use. Personal accounts are tested regularly, and Gemini Notebook Enterprise support is experimental.

A setup wizard detects the AI tools on the machine, connects the MCP server to them, and installs an agent skill that teaches the assistant how to use Gemini Notebook. Supported clients include Claude Code, Claude Desktop, Codex, Gemini CLI, Cursor, Windsurf, Cline, Antigravity, GitHub Copilot in VS Code, and OpenCode, and the skill also installs for Hermes and OpenClaw. Queries sent from the CLI or the MCP server persist to the notebook's chat history in the web interface.

  • Notebooks and sources: create, list, and share notebooks, and add URLs, text, Google Drive files, YouTube videos, and uploaded files as sources
  • Querying: ask questions of one notebook or across several notebooks, and list, view, or export chat sessions
  • Studio content: generate audio podcasts, video explainers, slide decks, briefing docs, flashcards, infographics, mind maps, and interactive reports
  • Artifact downloads: download a single generated artifact or every artifact from one or all notebooks
  • Research: start web or Google Drive research and import the sources it finds
  • Drive sync: detect stale Google Drive sources and sync them
  • Batch and pipelines: batch query, create, and delete operations, plus multi-step pipelines and tag-based notebook selection
  • Agent setup: a wizard connects the MCP server and installs the agent skill for detected AI tools, and a doctor command diagnoses problems
  • Profiles: named login profiles for several Google accounts at once, with optional encrypted login storage kept in the OS keychain
  • Session refresh: expired tokens and cookies refresh automatically through a headless browser when a login profile is saved

02 / Discussion CREDIBILITY-GATED

Discussion

Reading is open to everyone. Posting and voting need a verified identity or a GitHub grade of B or higher.

  • No discussions yet.

04 / Build

Build with Gemini Notebook MCP CLI.

Browse the catalog for frameworks, tools, and harnesses, each scored on real GitHub credibility.

Get Gemini Notebook MCP CLI →

Browse the catalog