Claude Skill MartBrowse skillsQuick linesLearn by videoTerminal guideWhat is a Skill?
 
← Back to list

NotebookLM Automation (notebooklm-py)

Operate Gemini Notebook (NotebookLM) end to end via the notebooklm-py CLI or typed async Python API — auth, sources, grounded chat, research, and artifact generation.

AutomationAdvanced★ 19,437⑂ 2,600AI score 9/10Last updated: Sep 23, 2026

What it does

  • Drives the notebooklm CLI (or the typed async Python client) to create notebooks, add sources, wait for readiness, run grounded chat, and generate/download audio, video, slide decks, infographics, reports, mind maps, data tables, quizzes, and flashcards.
  • Establishes reliable authentication: verify with notebooklm auth check --test --json, prefer durable profile-backed master tokens over cookie snapshots for headless/CI use, and treat both auth files as bearer credentials.
  • Enforces automation invariants: always --json, retain full UUIDs, pass -n <notebook_id> on every scoped command, and isolate concurrent agents with NOTEBOOKLM_PROFILE.
  • Defines authorization boundaries — deletions, share removal, logout, account-global language set, downloads, research wait --import-all, and note-creating commands require explicit user confirmation.
  • Ships a read-only failure-diagnosis routine plus exit-code semantics (0/1/2), timeout handling, and rate-limit guidance.

Who it's for

  • Researchers and content/data teams who want scripted NotebookLM workflows instead of manual clicking.
  • Engineers building document-summary or podcast pipelines on headless servers and CI.
  • Anyone running multiple agents concurrently who must keep credentials and context isolated.

Examples

  1. "Build a notebook from these three URLs and produce a podcast": create → source add ×3 → source wait → generate audio → artifact wait → download audio ./podcast.m4a.
  2. "Kick off deep research on topic X": source add-research "..." --mode deep --no-wait --json, keep the run ID, then import only when authorized with research wait --import-all.
  3. "Why did that command fail?": read-only triage with auth check --test --passive --json, source list, and artifact list before any state change.

· · · Install guide · · ·

Try it now, no install

Paste this into Claude to use the skill without installing anything.

Read the instructions in this file and follow them to help me:
https://raw.githubusercontent.com/teng-lin/notebooklm-py/HEAD/SKILL.md

What I want: (describe your task here)

If Claude can't open the link, open it yourself and paste the contents instead.

↓ If it works for you, download the ZIP below and install it. Then it runs on its own — no pasting each time.

Install in the Claude app (no terminal)
  1. Download the ZIP with the button below.
  2. In Claude, open Settings → Capabilities and turn on 'Code execution and file creation'. (one time)
  3. Go to Customize → Skills → + → 'Upload a skill' and upload the ZIP.
↓ Download ZIP
Install in Claude Code

Let Claude do it — paste this into Claude Code

Install the skill I found on Claude Skill Mart.
Copy the SKILL.m folder from the GitHub repo teng-lin/notebooklm-py into my ~/.claude/skills/notebooklm/.
When it's done, tell me in one line what this skill can do.

Install with a command instead

pip install "notebooklm-py[browser]" && notebooklm skill install

⚠ This is a third-party skill. Check the source repository before installing.

  1. Confirm Python 3.10+: python3 --version.
  2. Install into your existing environment: pip install "notebooklm-py[browser]". For CLI-only use, uv tool install "notebooklm-py[browser]" or pipx works well. If pip reports externally-managed-environment, avoid --break-system-packages and use uv/pipx or a virtualenv.
  3. Install the skill into your agent: notebooklm skill install (in sandboxed environments, build an archive with notebooklm skill package).
  4. Verify installation: notebooklm skill status --json.
  5. Authenticate: with a display, run notebooklm login. For unattended use, pip install "notebooklm-py[headless]", run notebooklm login --master-token --account <email> on a trusted workstation, and deploy only master_token.json to the target profile with mode 0600.
  6. Validate: notebooklm auth check --test --json must return status == "ok" and checks.token_fetch == true.
  7. Restart Claude Code and try a prompt like "list my NotebookLM notebooks" to confirm the skill triggers.
View source on GitHub ↗License: MIT