</>MCP Agents Market
Skill

notebooklm-py

by teng-lin18.8kPythonUpdated 2026-08-17

Unofficial Python API and agentic skill for Google Gemini Notebook. Full programmatic access to NotebookLM's features—including capabilities the web UI doesn't expose—via Python, CLI, and AI agents like Claude Code, Codex, and OpenClaw.

Claude CodeCodexClaude DesktopChatGPT

notebooklm-py is an unofficial Python library and agent skill that gives AI agents programmatic access to Google NotebookLM (now Gemini Notebook). It exposes the full NotebookLM feature set—including capabilities not available in the web interface—through Python APIs, CLI commands, an MCP server, and agent skill files for Claude Code and Codex. Developers use it to build research automation, content generation pipelines, grounded memory systems, and batch export workflows that leverage NotebookLM's AI-powered document analysis and artifact creation.

Key Features

Complete NotebookLM control via Python API, CLI, MCP server, and agent skills for Claude Code/Codex
Add sources from URLs, PDFs, YouTube, Google Drive, audio/video, and run automated web/Drive research with auto-import
Generate audio overviews (podcasts), videos, slide decks, quizzes, flashcards, infographics, data tables, mind maps, and reports
Batch downloads in multiple formats (MP3, MP4, PDF, PPTX, CSV, JSON, Markdown, HTML) with programmatic export
Grounded Q&A with source citations, conversation history, custom personas, and save-to-notes functionality
Multi-account profiles, master-token authentication for unattended workflows, and browser cookie import
Capabilities beyond web UI: bulk artifact downloads, quiz/flashcard export, mind map JSON extraction, individual slide revision
Remote MCP connector deployment for mobile access via claude.ai and ChatGPT with tunneling support

Use Cases

  • 01Zero-token research offload where NotebookLM handles document analysis and agents orchestrate with minimal token spend
  • 02Persistent cross-session memory for coding agents using notebooks as a grounded knowledge base with citations
  • 03Automated content repurposing from single source sets into podcasts, videos, slide decks, quizzes, and blog posts
  • 04Bulk export of study materials (flashcards, quizzes, mind maps) to Anki, Obsidian, or custom learning platforms
  • 05Scheduled audio briefing generation for personalized podcast feeds from research notebooks
  • 06Incident runbook automation that generates diagnostic briefings from technical documentation on alerts

Related Skills

View more

notebooklm-py — FAQ

What is notebooklm-py and what does it do?+

notebooklm-py is an unofficial Python library and agent skill that provides programmatic access to Google NotebookLM (Gemini Notebook). It lets AI agents and developers control notebooks, add sources, ask questions, generate audio/video/quiz/flashcard content, and export artifacts through Python, CLI, or MCP server interfaces.

How do I install notebooklm-py for use with Claude Code?+

Install with uv tool or pipx: `uv tool install "notebooklm-py[browser]"`, then authenticate with `notebooklm login`. For agent integration, run `notebooklm skill install` to add the skill to Claude Code's skills directory, or use `npx skills add teng-lin/notebooklm-py`.

Which AI clients and agents work with notebooklm-py?+

It works with Claude Code, Codex, Claude Desktop (via MCP server), ChatGPT (with remote MCP connector), and any AI agent that can use MCP servers or skill files. The library also provides a standalone Python API and CLI for custom integrations.

Does notebooklm-py require API keys or special credentials?+

You need a Google account with access to NotebookLM/Gemini Notebook. Authentication uses browser login (Playwright), imported browser cookies, or master tokens. No separate API keys are required, but it uses undocumented Google APIs that may change without notice.

Is notebooklm-py free to use?+

The library itself is free and open-source (MIT license). However, it operates within your Google account's NotebookLM quota limits, which vary by tier (free vs. paid Google accounts). The library does not incur additional costs beyond your existing Google account.

What formats can I export from NotebookLM using this library?+

You can download audio (MP3/M4A), video (MP4), slide decks (PDF/PPTX), quizzes and flashcards (JSON/Markdown/HTML), data tables (CSV), mind maps (JSON), infographics (PNG), and reports (Markdown). Many of these exports are unavailable or limited in the web interface.

How do I install notebooklm-py?+

Open the source repository on GitHub and follow its README. notebooklm-py is a skill — MCP Agents Market links you directly to the official repo.

Is notebooklm-py free?+

notebooklm-py is an open-source project hosted on GitHub. Check the repository for its license and any usage requirements.

Related searches