ClawHub
Back to skill

Security audit

Evolver

Security checks across malware telemetry and agentic risk

Overview

This skill contains no executable code, but it tries to make the agent adopt an always-on, self-updating execution persona without clear limits or reset controls.

Review before installing. Use it only for clearly bounded missions, keep normal safety and stop instructions in force, require confirmation before tool use that affects external systems or money, and do not allow permanent memory or behavior-rule updates unless you have explicit controls to review and reset them.

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
90% confidence
Finding
The activation phrases are broad enough to match ordinary requests like 'make this happen' or 'turn this into a business,' which can cause the skill to engage without clear user intent. Because the skill also declares an always-active identity and mandatory response format, accidental activation could override normal agent behavior across unrelated tasks.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.