ClawHub
Back to skill

Security audit

NotebookLM Ops

Security checks for vulnerabilities and agentic risk

Overview

The skill appears purpose-built for NotebookLM authentication automation, but it controls an authenticated Google browser session through VNC/CDP and delegates to unbundled local scripts without enough scoping detail.

Review before installing. Use a dedicated low-privilege Google account or browser profile, confirm VNC and CDP bind only to trusted local interfaces with appropriate access controls, and inspect the /home/moltuser/clawd scripts this skill delegates to before running it.

A.I.G

Vulnerability Patterns
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs

SkillSpector

Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Behavioral ASTexec() Call, eval() Call, Dynamic Import
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (4)

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The skill states it will start Xvfb, openbox, x11vnc, Chromium, navigate to NotebookLM, and refresh auth via CDP, but it does not clearly warn that it launches remote-view/control-capable GUI services and performs browser automation inside the user's authenticated session. In this context, hidden GUI/VNC startup plus CDP-driven browser control materially expands attack surface and could enable session observation or manipulation if exposed or misconfigured.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The skill explicitly relies on reusing a persistent Chromium profile and Google session cookies, but it does not clearly warn the user that long-lived authenticated state will be retained and reused for automated actions. This increases the risk of unintended access, session misuse, and exposure of authenticated Google data if the host, profile, or automation stack is accessed by another local user or process.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
94% confidence
Finding

The trigger list is written only in Russian, which imposes a specific language for invocation. Under the policy, language constraints should be optional, user-selectable, or clearly documented as a justified locale-specific limitation.

Content

No source excerpt is available for this finding.

Missing User Warnings

Low
Category
Not specified by scanner
Confidence
88% confidence
Finding

This shell script invokes other scripts, which is a safety-relevant operation for code files, but the file contains no comment, echo/log statement, or other user-facing disclosure about what will be run. A user reading or invoking this wrapper gets no warning that it will perform authentication-status and listing actions via subprocess execution.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.