Skip to content
Get packs

Gemini Notebook Automation

Lets an agent install, authenticate, and operate Google's Gemini Notebook via the notebooklm-py CLI or Python API.

Source Download skill
Repository
teng-lin/notebooklm-py
Stars
20,066
Licence
MIT
Files
9 files

What it does

It teaches the agent how to drive Gemini Notebook (Google's AI notebook tool) from the command line or Python: creating notebooks, adding sources, asking grounded questions, generating artifacts like mind maps, and downloading results. It also gives safety rules: check that login actually works before starting, pin commands to exact IDs so concurrent work does not collide, and ask the user before deleting, downloading, or generating anything long-running. The skill relies on an unofficial package, notebooklm-py, that talks to the service on the user's behalf.

Great for

  • Developers automating Gemini Notebook workflows in scripts or CI
  • Researchers who want an agent to add sources and pull grounded answers from a notebook
  • Users who already have the notebooklm-py CLI set up and want reliable agent-driven operation

Skip it if

  • Users who only want the official Gemini API: the skill itself says not to use it for that, and it is unrelated to the API
  • Unattended automation using a master token on shared or untrusted machines: a master token is a full-account credential that survives password changes, so a leak exposes the whole account
  • Users who cannot tolerate their Google session cookies or tokens being handled by a third-party tool

Try asking

  • “Create a notebook called 'Q3 research', add these three URLs as sources, wait until they are ready, then ask for a summary”
  • “Generate a mind map from my notebook abc123 and tell me when it is done”
  • “Check whether my notebooklm login still works before we start the batch job”
  • “Download the audio overview artifact with this exact ID from notebook xyz789”
Step-by-step guide: how to use Gemini Notebook AutomationInstall, first run, examples, tips and FAQs

Install in 3 steps

  1. 1

    Download the skill

    The ZIP contains the 9 original files, the MIT licence and an attribution note.

    notebooklm.zip
  2. 2

    Add it to your agent

    Claude Code (all projects)

    mkdir -p ~/.claude/skills && unzip notebooklm.zip -d ~/.claude/skills/

    Claude Code / Agent SDK (this project only)

    mkdir -p .claude/skills && unzip notebooklm.zip -d .claude/skills/

    Claude apps (web/desktop): Settings → Capabilities → Skills → Upload skill, and choose the ZIP.

  3. 3

    Ask for the task

    No special command needed — the agent reads the skill's description and loads it when your request matches. Try one of the prompts above.

Get better results

  • Run 'notebooklm auth check --test --json' first and only proceed when status is ok and token_fetch is true
  • Keep auth files out of logs and git, set them to mode 0600, and use a dedicated account for any master token
  • Use --json and retain full IDs from each command rather than relying on the 'latest' item or 'notebooklm use'

What's inside

  • .pre-commit-config.yaml
  • AGENTS.md
  • CLAUDE.md
  • CONTRIBUTING.md
  • hatch_build.py
  • pyproject.toml
  • README.md
  • SECURITY.md
  • SKILL.md

Redistributed unmodified from teng-lin/notebooklm-py under the MIT licence. The assessment was drafted with AI from the skill's files on 2026-10-29. Skills can instruct your agent to run code — read SKILL.md and any scripts before installing.