ClawHub

Auto Content

Security checks across malware telemetry and agentic risk

Overview

This SEO content skill is mostly coherent, but it tells the agent to automatically run web searches using project details before the user asks.

Review before installing. Use it only if you are comfortable with SEO research workflows that search the web, and disable or require confirmation for automatic Stage 0 trend scanning. Avoid putting confidential launch plans, strategy, private competitor analysis, or unrestricted API keys in the project spec.

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 (3)

Vague Triggers

High
Confidence
97% confidence
Finding
The skill’s trigger conditions are overly broad and explicitly instruct automatic activation at the start of every session, even before the user asks for content help. This can cause unsolicited external searches and processing of project context, violating least-astonishment and potentially exposing user data or browsing intent without clear consent.

Vague Triggers

High
Confidence
99% confidence
Finding
Stage 0 is configured to run automatically on every new session before any user message, creating an ambiguous and non-consensual activation path. This is dangerous because it can initiate external web searches using project data or inferred context without user awareness, leading to privacy leakage and unnecessary external actions.

Missing User Warnings

Medium
Confidence
93% confidence
Finding
The skill description says it will scan X for trending topics automatically but does not prominently warn the user that proactive web searches and external data access may occur before user input. Lack of disclosure undermines informed consent and increases the risk of unexpected data handling or external query generation tied to the user’s project.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal