notebooklm

1# Gemini Notebook Automation
2 
3Use the `notebooklm` CLI for agent workflows. Prefer `--json` and explicit IDs so every
4operation is inspectable and safe under concurrency. Use the typed async Python API only when the
5user requests application code or the CLI cannot express the workflow. The readiness, identity,
6authorization, and credential-handling rules below apply to both interfaces.
7 
8## Setup and Authentication
9 
10Requires Python 3.10+. Install the package in the user's existing environment; do not create a
11separate environment unless requested:
12 
13```bash
14pip install "notebooklm-py[browser]"
15pip install "notebooklm-py[cookies]"  # optional browser-cookie extraction
16```
17 
18If system `pip` reports `externally-managed-environment`, do not use `--break-system-packages`.
19For CLI-only use, offer `uv tool install "notebooklm-py[browser]"` or the equivalent `pipx`
20command; for application code, use the user's active project environment.
21 
22For unattended or headless work, prefer durable profile-backed master-token auth over a copied
23cookie snapshot. Install `pip install "notebooklm-py[headless]"`; the one-time automatic OAuth
24capture also needs `[browser]`. On a trusted workstation run
25`notebooklm login --master-token --account <email>`, then deploy `master_token.json`, not
26`storage_state.json`, to the selected profile. `NOTEBOOKLM_HOME` selects the private base directory
27and `NOTEBOOKLM_PROFILE` selects its profile; defaults resolve to
28…