notebooklm

Automates Google NotebookLM notebooks, sources, and artifact generation via CLI.

Updated May 13, 2026
One-click install
npx skills add https://github.com/Eduardo-Sebastian-Paipay-Vega/Documentaci-n_de_dise-o_de_Sistemas_y_Modulos --skill notebooklm-eduardo-sebastian-paipay-vega
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: notebooklm
Source: https://github.com/Eduardo-Sebastian-Paipay-Vega/Documentaci-n_de_dise-o_de_Sistemas_y_Modulos/tree/main/GIMNASIO/.claude/skills/notebooklm
Command: npx skills add https://github.com/Eduardo-Sebastian-Paipay-Vega/Documentaci-n_de_dise-o_de_Sistemas_y_Modulos --skill notebooklm-eduardo-sebastian-paipay-vega

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires notebooklm-py.

What problem does it solve? Google NotebookLM's web UI requires manual clicking for every notebook, source upload, and artifact generation, making batch research workflows and automation impossible. This Skill provides full programmatic CLI access to NotebookLM, including capabilities not exposed in the web interface. ## Core Features & Use Cases - Notebook and Source Management: Create notebooks and add sources from URLs, YouTube, PDFs, audio, video, and images, with deep web research modes that auto-import sources. - Artifact Generation and Download: Generate podcasts, videos, slide decks, infographics, reports, quizzes, flashcards, mind maps, and data tables, then download them as MP3, MP4, PDF, PPTX, CSV, JSON, or Markdown. - Chat with Citations: Query notebook content programmatically and receive answers with source references, with options to save conversations as notes. - Use Case: A researcher collects 20 URLs on a topic, adds them to a notebook, waits for indexing, generates a podcast episode, and downloads the MP3—all through scripted commands with background subagents handling long waits. ## Quick Start Run notebooklm login to authenticate, then ask the agent to create a notebook, add your source URLs, and generate a podcast about your topic.

Frequently Asked Questions about notebooklm

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I create a NotebookLM podcast from the command line?

Create a notebook with notebooklm create, add sources with notebooklm source add, wait for them to reach ready status, then run notebooklm generate audio with optional instructions. Download the result with notebooklm download audio once the artifact completes.

How do I authenticate the NotebookLM CLI?

Run notebooklm login to open a browser for Google OAuth, then verify with notebooklm list. For CI/CD environments, set the NOTEBOOKLM_AUTH_JSON environment variable with your storage_state.json contents instead of interactive login.

What source types does NotebookLM support?

NotebookLM accepts PDFs, YouTube URLs, web URLs, Google Docs, text files, Markdown, Word documents, EPUB, audio files, video files, and images. Source limits vary by plan, from 50 on Standard up to 600 on Ultra.

Can I run multiple NotebookLM agents in parallel?

Yes, but avoid shared context conflicts by passing explicit notebook IDs with -n or --notebook flags. Alternatively, isolate agents using separate NOTEBOOKLM_PROFILE or NOTEBOOKLM_HOME environment variables per agent.

Why does NotebookLM audio or video generation fail?

Generation failures are usually caused by Google rate limits, not CLI bugs. Check status with notebooklm artifact list, wait 5-10 minutes, and retry; audio typically takes 10-20 minutes and video 15-45 minutes to complete.

What can the NotebookLM CLI do that the web UI cannot?

The CLI supports batch downloads of all artifacts, quiz and flashcard export as JSON/Markdown/HTML, mind map extraction as JSON, slide decks as editable PPTX, individual slide revision, source fulltext retrieval, and programmatic sharing management.