ClawHub
Back to skill

Security audit

Content Channel Research

Security checks across malware telemetry and agentic risk

Overview

This is a text-only content research checklist with no code or private-data access; the main caveat is that it may trigger a heavier research workflow than some users expect.

Install this if you want an assistant to run structured public research before committing to a video or content topic. Be aware it may invoke a full multi-phase workflow for broad content-planning prompts, so explicitly tell the agent to skip or shorten the workflow when you only want brainstorming, scripting, or quick advice.

SkillSpector

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

Vague Triggers

Medium
Confidence
87% confidence
Finding
The trigger language is broad enough to activate on common content-planning questions, which can cause the agent to invoke this skill when the user did not ask for a full research workflow. That creates scope creep, unnecessary external-search behavior, and may override more appropriate or narrower skills, reducing reliability and potentially exposing more user context than needed.

Vague Triggers

Medium
Confidence
91% confidence
Finding
Labeling the skill as 'Mandatory before' several broad creator tasks encourages unconditional invocation without checking whether the workflow is necessary, proportional, or desired by the user. In agent systems, this kind of expansive activation guidance can lead to overreach, wasted effort, and incorrect routing of routine requests into a heavier process.

VirusTotal

58/58 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.