Skill Detail

Search NotebookLM notebooks, sources, and generated study artifacts from agents

<p>Use notebooklm-mcp-cli when an agent needs to search NotebookLM notebooks, add sources, run notebook queries, and retrieve generated study artifacts without leaving an MCP workflow.</p>

Research & ScrapingMCP
Research & Scraping MCP Security Reviewed
⭐ 3.6k GitHub stars
INSTALL WITH ANY AGENT
npx skills add agentskillexchange/skills --skill search-notebooklm-notebooks-sources-and-generated-study-artifacts-from-agents Copy
Works best when you want a reusable capability, not another fragile one-off prompt.
At a glance
Tools required
A NotebookLM account, browser-based authentication or cookie import, Python tooling via uv, pip, or pipx, and an MCP-compatible client if you want agent-side access.
Install & setup
<p>Install the unified package with <code>uv tool install notebooklm-mcp-cli</code>, <code>pip install notebooklm-mcp-cli</code>, or <code>pipx install notebooklm-mcp-cli</code>. Run <code>nlm login</code> to authenticate, then configure your client with <code>nlm setup add claude-code</code>, <code>nlm setup add cursor</code>, <code>nlm setup add gemini</code>, or point it directly at the <code>notebooklm-mcp</code> executable.</p>
Author
jacob-bd
Publisher
Individual
Last updated
Apr 14, 2026
Quick brief

notebooklm-mcp-cli gives an agent programmatic access to Google NotebookLM through an MCP server and companion CLI. It can list or create notebooks, add URLs, files, Drive items, or text as sources, run notebook queries that persist into NotebookLM, kick off research or studio artifacts such as audio and slides, and pull results back into a broader research workflow.

How it works

What this skill actually does

The scope boundary is clear enough to be skill-shaped: this is for agent-side NotebookLM research operations, not a generic Google product card, not a broad browser automation workaround, and not a general note-taking platform listing. Invoke it when an agent needs NotebookLM’s source-grounded notebook workflows inside a repeatable research process, not when a human can just use the NotebookLM web UI directly.