notebooklm-mcp-cli

What it is and what it does

notebooklm-mcp-cli is a unified command-line interface and Model Context Protocol server that bridges your terminal and AI agents to Google Gemini Notebook. It lets you programmatically create notebooks, add sources (URLs, files, Google Drive), generate podcasts and other studio content, query notebooks, manage chat sessions, and download artifacts—all from the CLI or by connecting AI assistants like Claude or Cursor as MCP clients.

The package wraps Gemini Notebook's internal APIs (which are undocumented and subject to change) and provides two entry points: the `nlm` CLI for direct terminal use and scripting, and the `notebooklm-mcp` server for AI agents to call as tools. It supports batch operations, cross-notebook queries, pipelines for multi-step workflows, and tagging. Setup is straightforward for Claude Desktop (one-click `.mcpb` extension) or manual for other tools via JSON config generation.

Use it for:

• Automate research workflows: create notebooks, add sources, and generate podcasts or summaries without manual UI interaction.
• Integrate Gemini Notebook into AI agent workflows: connect Claude, Cursor, or other MCP-compatible tools to let them manage notebooks on your behalf.
• Batch process multiple notebooks: query, create, or delete notebooks in bulk via CLI commands or pipeline definitions.
• Download and archive artifacts: export all generated content (audio, slides, videos) from one or many notebooks to local storage.
• Extend CI/CD pipelines: trigger notebook creation and content generation as part of automated research or documentation workflows.

Worth the install?

AI-flagged interpretation of the facts on this page — verify before relying

Provides CLI and MCP server access to Google Gemini Notebook (formerly NotebookLM), enabling programmatic notebook creation, source management, content generation, and AI agent integration.

Yes. The package is actively maintained (released 1 day ago), has low install friction, carries no known vulnerabilities, and uses a permissive MIT license. It fills a genuine gap for programmatic Gemini Notebook access. The main caveat is that it relies on undocumented internal APIs that may break without notice—suitable for personal automation and experimentation, but use with caution in production workflows where API stability is critical.

Install

notebooklm-mcp-cli on PyPI

pip install notebooklm-mcp-cli

uv add notebooklm-mcp-cli

poetry add notebooklm-mcp-cli

Installing notebooklm-mcp-cli

Before you install

Low friction install with a pure-Python wheel and active maintenance—released 1 day ago with 5852 GitHub stars. Nine runtime dependencies are all common, well-maintained packages (fastmcp, httpx, pydantic, typer, etc.), suggesting a stable dependency footprint.

License in practice

MIT license (permissive) means you can use, modify, and distribute this package freely with minimal restrictions, making it suitable for both personal and commercial projects.

Quickstart

pip install notebooklm-mcp-cli
nlm notebook list
nlm notebook create "My Research"
nlm source add --url "https://example.com"
nlm studio create audio

Requires Python 3.11 or later. Authentication requires extracting cookies from your browser session with Gemini Notebook; the package provides tooling for this but it is a manual setup step.

Verify before relying

• Whether all Gemini Notebook features are fully supported or if some remain untested (description notes testing with Pro/free and Ultra tiers, but Enterprise support is unverified).
• Stability guarantees for internal API usage—the package relies on undocumented Gemini Notebook APIs that may change without notice.
• Whether the MCP server works reliably with all advertised AI tools (Claude, Gemini, Cursor, GitHub Copilot, Cline) or if some integrations have known limitations.

Package facts

Evidence: notebooklm_mcp_cli-0.9.11-py3-none-any.whl

Keywords: ai, cli, google, mcp, notebooklm, podcast, research

Tags

More Artificial Intelligence packages