ClawHub
Back to skill

Security audit

Notebooklm Youtube Skill

Security checks for vulnerabilities and agentic risk

Overview

This skill does what it says: it uses Chrome, web search, YouTube, and NotebookLM to create account-linked notebooks and audio summaries from user-provided videos.

Install only if you are comfortable with the agent controlling Chrome, using your logged-in NotebookLM/Google session, searching the web for people in the video, and creating persistent NotebookLM notebooks/audio. Avoid using it with confidential, regulated, private, or copyrighted material unless you have permission and accept the third-party data sharing.

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
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • 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
Findings (4)

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
96% confidence
Finding

The README advertises automated web research, NotebookLM notebook creation, transcript ingestion, and audio generation in a user account without warning about privacy implications or irreversible external actions. Users may supply sensitive or copyrighted videos without understanding that the skill will send data to third-party services and create persistent artifacts in their Google account.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The casual invocation phrase is broad enough that ordinary user requests like 'Create a NotebookLM notebook for this video' could trigger the skill unintentionally. Because the skill performs browser automation and creates content in a logged-in NotebookLM account, accidental activation can cause external side effects the user did not explicitly authorize in that moment.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
97% confidence
Finding

The skill automates extraction of information from YouTube and then sends both the video URL and compiled research text to third-party services, including search engines and NotebookLM, but the description does not warn users that potentially sensitive or copyrighted content will be transmitted externally. This creates a meaningful privacy and compliance risk because users may invoke the skill without understanding that derived personal data and research content leave the local environment.

Content

No source excerpt is available for this finding.

Autonomous Decision Making

Medium
Category
Excessive Agency
Confidence
75% confidence
Finding

Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.

Content

Scanner excerpt · SKILL.md (reported line 173)May include surrounding context.

md
| ❌ Don't | ✅ Do |
|----------|-------|
| `document.querySelector('textarea')` | `textarea[placeholder="Paste any links"]` |
| Click without checking state | Verify button enabled first |
| Assume action succeeded | Screenshot to confirm |
| Rush through steps | Wait for async operations |

Static analysis

No suspicious patterns detected.