This is an experimental CLI wrapper that lets Claude interact with Google's NotebookLM through the unofficial notebooklm-py library. It exposes commands for managing notebooks, adding sources, asking questions, generating artifacts like study guides or FAQs, and downloading the outputs. You'd use this if you want Claude to orchestrate NotebookLM workflows without manual clicking, like batch processing documents or chaining questions across multiple notebooks. The harness follows the CLI-Anything pattern, so it's structured for agent use with JSON output flags where supported. Just know it's unofficial, relies on browser automation under the hood via Playwright, and requires a valid local NotebookLM session to work.
npx -y skills add hkuds/cli-anything --skill cli-anything-notebooklm --agent claude-codeInstalls into .claude/skills of the current project.
Select a file.
sentry/dev
degausai/wonda
github/awesome-copilot
insforge/agent-skills
github/awesome-copilot
breaking-brake/cc-wf-studio