A powerful CLI tool and AI agent skill providing full programmatic access to Google NotebookLM features, including advanced capabilities not available in the web UI.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install notebooklm-api
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install notebooklm-api using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is NotebookLM Automation Skill?
The NotebookLM Automation skill empowers developers and AI agents to interact with Google NotebookLM entirely through code or command-line interfaces. It bridges the gap between manual research and automated content pipelines by allowing users to manage notebooks, upload diverse source types (PDFs, YouTube, URLs, and images), and interact with indexed content through a specialized chat API.
Beyond basic management, this Openclaw Skills integration unlocks high-fidelity content generation. You can programmatically trigger the creation of AI-generated podcasts, video explainers, quizzes, and mind maps. By utilizing structured JSON outputs and background subagents, users can build complex research workflows that operate autonomously, bypassing the limitations of the standard web interface.
NotebookLM Automation Skill Use Cases
- Automated Podcast Creation: Generate audio overviews from a list of research URLs or documents without manual clicks.
- Deep Web Research: Use the deep research mode to find, analyze, and import web sources on specific topics automatically.
- Bulk Document Processing: Upload dozens of PDFs or Google Docs to a notebook for instant summarization and structured data extraction.
- Educational Content Generation: Automatically transform study materials into flashcards, quizzes (exported as Markdown), and visual mind maps.
- CI/CD Integration: Incorporate notebook analysis or content generation into automated pipelines using service account-like authentication.
How NotebookLM Automation Skill Works
- Authentication: The user authenticates via
notebooklm login, which performs a Google OAuth flow to establish a secure session. - Context Selection: A notebook is created or selected using
notebooklm use, setting the working environment for subsequent commands. - Source Ingestion: Documents, URLs, or YouTube videos are added to the notebook; the system then indexes these items for retrieval-augmented generation.
- Task Execution: The user issues commands for specific intents, such as
askfor querying the knowledge base orgeneratefor creating complex artifacts like audio or reports. - Status Monitoring: For long-running tasks like video or audio generation, the skill tracks progress via artifact IDs and status polling.
- Export & Delivery: Once processing is complete, the structured data (CSV, JSON, Markdown) or media files (MP3, MP4) are downloaded to the local filesystem.
NotebookLM Automation Skill Setup
First, install the core package from PyPI:
pip install notebooklm-py
Next, install the specific AI agent skill to enable Openclaw Skills functionality:
notebooklm skill install
Finally, authenticate your session:
notebooklm login
notebooklm status
NotebookLM Automation Skill Data Schema & Taxonomy
The skill organizes data around the concept of Notebooks and Artifacts. The following table describes the primary data entities:
| Entity | Description | Format |
|---|---|---|
| Notebook | The primary container for projects. | JSON with UUID, Title, Created Date |
| Source | The raw input data (URLs, Files, etc.). | Indexed text/Fulltext with status (READY/ERROR) |
| Artifact | Generated content (Audio, Video, PDF). | MP3, MP4, PDF, MD, CSV, or JSON |
| Context | Local state for the current active notebook. | ~/.notebooklm/context.json |
| References | Citation data linked to specific chat answers. | JSON containing source_id and cited_text |
NotebookLM Automation Skill Advanced Features
- Subagent Parallelism: Supports isolated environments via
NOTEBOOKLM_HOMEfor running multiple agents concurrently without context collisions. - Extended Export Formats: Export quizzes and flashcards as JSON or Markdown, a feature exclusive to this skill and not found in the web UI.
- Deep Research Mode: An autonomous search capability that performs comprehensive web analysis and imports findings directly into a notebook.
- Global Language Controls: Programmatically set and switch between 80+ supported languages for all AI-generated content.
- JSON API Access: Use the
--jsonflag on almost any command to pipe structured data into other CLI tools or scripts.
SKILL.md
Loading
Related Openclaw Skills
Gasless Operations Analyzer
A comprehensive CLI tool designed to analyze, track, and manage blockchain gasless operations and protocol security concepts.
SPX Express Tracking
A specialized logistics skill for querying real-time SPX Express and Cainiao Malaysia shipment data with advanced bottleneck analysis.
De-AI Writing (Chinese)
A specialized skill designed to strip away the robotic, formulaic nature of AI-generated Chinese text to produce natural, professional, and high-density content.
ClawSpa
ClawSpa is a comprehensive wellness and maintenance suite designed to optimize AI agent performance through memory cleanup, security audits, and alignment diagnostics.
Amazon Supplier Sourcing
An end-to-end AI sourcing co-pilot designed to help Amazon FBA sellers find, evaluate, and negotiate with global manufacturers.
Dida365 (TickTick) Task Management Skill
A comprehensive task and project management skill for Dida365 and TickTick that enables AI agents to create, track, and organize workflows via a high-performance CLI.
Featured*
