What problem does it solve? Researching blog content often requires grounding claims in your own source documents, but manually searching Google NotebookLM notebooks and copying answers into your writing workflow is slow and disconnected. This Skill lets you query NotebookLM directly from Claude Code and receive source-grounded, citation-backed answers without leaving your editor. ## Core Features & Use Cases - Notebook Querying: Ask questions against any NotebookLM notebook via headless browser automation and receive answers grounded in your uploaded documents. - Notebook Library Management: Add, search, activate, and remove notebooks with metadata (name, description, topics) stored locally in a JSON library. - Authentication & Session Handling: One-time interactive Google login with persistent browser profile and cookie injection, plus status checks and re-authentication. - Internal API for Blog Workflows: blog-write and blog-researcher can call it as a subagent for research context, with silent graceful fallback to WebSearch when unauthenticated. - Use Case: While drafting a blog post about marketing benchmarks, run /blog notebooklm ask "What are the latest conversion rate benchmarks?" against your research notebook and cite the underlying sources directly in the article. ## Quick Start Run /blog notebooklm setup to authenticate with your Google account, then ask a question with /blog notebooklm ask followed by your research question.
