nlm-skill

Automates NotebookLM notebooks, sources, research, and content generation via the nlm CLI and MCP server.

Updated Jul 19, 2026
One-click install
npx skills add https://github.com/DagimAlemayehuu/AgenticEngineering --skill nlm-skill-dagimalemayehuu
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: nlm-skill
Source: https://github.com/DagimAlemayehuu/AgenticEngineering/tree/main/agentic-engineering/core/dependencies/research/nlm-skill
Command: npx skills add https://github.com/DagimAlemayehuu/AgenticEngineering --skill nlm-skill-dagimalemayehuu

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve? Interacting with Google NotebookLM programmatically requires knowing dozens of CLI commands, MCP tool signatures, authentication flows, and confirmation rules. This Skill provides complete operational guidance so an AI agent can manage notebooks, sources, research, and content generation without trial and error. ## Core Features & Use Cases - Notebook and Source Management: Create notebooks, add sources from URLs, YouTube, text, files, and Google Drive, and sync stale Drive documents. - Research and Content Generation: Run fast or deep web research, import discovered sources, and generate podcasts, reports, quizzes, flashcards, mind maps, slides, infographics, videos, and data tables. - Batch and Cross-Notebook Operations: Query multiple notebooks at once, tag notebooks, run pipelines, and perform batch actions across tagged sets. - Use Case: Ask the agent to research a topic, import the discovered sources into a new notebook, and generate a deep-dive podcast episode, all through guided nlm commands. ## Quick Start Use the nlm-skill to create a NotebookLM notebook, add these URLs as sources, and generate an audio overview.

Frequently Asked Questions about nlm-skill

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

FAQPage Schema
How do I generate a podcast from NotebookLM sources using the CLI?▼

Run nlm audio create with the notebook ID and the --confirm flag, optionally specifying a format like deep_dive, brief, critique, or debate. Poll nlm studio status to track generation, which typically takes two to five minutes.

How do I add sources to a NotebookLM notebook programmatically?▼

Use nlm source add with the notebook ID and a source flag: --url for web pages or YouTube, --text with --title for pasted content, --drive for Google Drive documents, or --file for local files like PDFs and images.

Should I use the NotebookLM MCP tools or the nlm CLI?▼

Both share the same authentication backend and capabilities. If both are available, ask the user which they prefer; MCP tools are called directly with parameters like confirm=True, while the CLI runs via Bash with --confirm flags.

Why do nlm commands fail with authentication errors?▼

NotebookLM sessions expire after roughly 20 minutes, causing errors like 'Cookies have expired'. Re-run nlm login to re-authenticate, and use nlm login --check to validate the session before critical operations.

Can I query multiple NotebookLM notebooks at once?▼

Yes, use nlm cross query with --notebooks, --tags, or --all to get aggregated answers with per-notebook citations. The MCP equivalent is the cross_notebook_query tool with notebook_names, tags, or all parameters.

What are the limitations of NotebookLM automation with nlm?▼

Free tier accounts face roughly 50 queries per day and rate limits requiring delays between operations. The interactive chat REPL cannot be controlled by AI tools, so one-shot queries via nlm notebook query must be used instead.