ClawHub
Back to skill

Security audit

Profile Optimization

Security checks across malware telemetry and agentic risk

Overview

This skill is an advisory profile-optimization guide that does not install code, persist data, or directly change user accounts.

Installers should understand that this skill may activate for broad bio/profile requests and will use brand, audience, voice, and strategy files to tailor recommendations. It should be treated as copy and strategy guidance that the user applies manually, not as a tool that changes social accounts directly.

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 (1)

Vague Triggers

Medium
Confidence
94% confidence
Finding
The trigger description includes very broad phrases such as "profile" and common requests like "write my bio," which can cause the skill to activate in situations far beyond profile optimization. Over-broad invocation increases the chance of unintended context capture, inappropriate skill routing, and downstream actions based on irrelevant or insufficient user intent.

VirusTotal

62/62 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.