L’interface est localisée ; les présentations et documents des Skills sont traduits automatiquement par IA dans votre langue. Le code, les commandes, chemins et termes techniques restent en version originale, et le document de l’auteur fait foi.
Explorer les Skills
Productivité · SKILL.md

cli-anything-notebooklm

cli-anything-notebooklm

Experimental NotebookLM harness for listing notebooks, managing sources, asking questions, generating artifacts, and downloading outputs through an installed notebooklm CLI.

Independent GitHub snapshot · Automated classification · Original documentation governs

Hhkuds / cli-anythingApache-2.0

Documentation de la Skill

GitHub

name: cli-anything-notebooklm description: Experimental NotebookLM harness for listing notebooks, managing sources, asking questions, generating artifacts, and downloading outputs through an installed notebooklm CLI.

cli-anything-notebooklm

Experimental NotebookLM harness for CLI-Anything.

Installation

This package is intended to be installed from the harness directory:

cd notebooklm/agent-harness
python3 -m pip install -e .

Install the upstream NotebookLM CLI if needed:

python3 -m pip install --user 'notebooklm-py[browser]'
python3 -m playwright install chromium

Requirements

  • notebooklm command installed locally
  • Valid local NotebookLM login session

Usage

Basic Commands

# Show help
cli-anything-notebooklm --help

# Start with a notebook context
cli-anything-notebooklm --notebook nb_123 source list

# Prefer JSON for agent use
cli-anything-notebooklm --json notebook list

Command Groups

GroupPurpose
authlogin and auth validation
notebooknotebook list, create, summary
sourcesource listing and URL add
chatask questions and inspect history
artifactlist and generate artifacts
downloadfetch generated outputs
shareinspect sharing state

Agent Workflow

  1. Check auth with cli-anything-notebooklm auth status
  2. Discover notebook IDs with cli-anything-notebooklm --json notebook list
  3. Use explicit --notebook for follow-up commands
  4. Prefer --json only where the upstream notebooklm command supports it

Agent Guidance

  • Prefer explicit notebook IDs with --notebook.
  • Use --json for machine-readable output only on commands that support it upstream.
  • Treat this harness as experimental and unofficial.
  • Do not expose auth files or cookies in logs.
  • NotebookLM is a Google product; this harness is unofficial and not affiliated with Google.

References

Skills enregistrées