ClawHub

NotebookLM API

Security checks across malware telemetry and agentic risk

Overview

This appears to be a legitimate NotebookLM automation skill, but it can automatically send documents or prompts to Google NotebookLM under broad activation rules, so it needs careful review before installation.

Install only if you trust notebooklm-py and are comfortable granting it access to your Google NotebookLM account. Prefer explicit /notebooklm use, require confirmation before uploading local or sensitive documents, keep NOTEBOOKLM_AUTH_JSON out of logs and chat transcripts, and use a dedicated NOTEBOOKLM_HOME or account for automation.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (2)

Vague Triggers

Medium
Confidence
88% confidence
Finding
The skill's intent-based activation examples are broad enough to match ordinary requests such as summarization, quiz generation, or podcast creation without a clear requirement that the user explicitly wants NotebookLM. That can cause the agent to invoke an external service, create remote notebooks, or upload sources when the user only asked for a generic task, leading to unintended data disclosure and unexpected side effects.

Missing User Warnings

Medium
Confidence
93% confidence
Finding
The skill instructs users to authenticate with Google OAuth and to place `storage_state.json` contents into `NOTEBOOKLM_AUTH_JSON`, but it does not include strong guidance to treat these values as secrets, avoid logging them, avoid echoing them in commands, and prevent exposure in agent output. Because these materials likely contain active session state or cookies, mishandling them could let an attacker reuse credentials and access the user's NotebookLM account.

VirusTotal

59/59 vendors flagged this skill as clean.

View on VirusTotal