ClawHub

Zhua Evolver

Security checks across malware telemetry and agentic risk

Overview

This skill is not clearly harmful, but it asks an agent to automatically expand and change its own capabilities without enough approval controls.

Review before installing. Only use it under close supervision, and do not allow automatic skill searches, installations, configuration changes, continuous evolution, identity-file edits, or rollback-prevention behavior unless each action is shown first, sourced from a trusted registry, and easy to undo.

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

Medium
Confidence
88% confidence
Finding
The invocation guidance is overly broad, allowing the skill to be triggered for vague goals like becoming 'more intelligent' or 'continuous evolution' without clear boundaries or approval requirements. In an agent setting, this increases the chance of unnecessary or unsafe activation, especially because the skill also describes searching for and installing additional skills autonomously.

Missing User Warnings

High
Confidence
97% confidence
Finding
The markdown explicitly advertises automatic installation and configuration of new skills, which is a system-changing behavior, but provides no warning, approval gate, trust policy, or validation requirements. In this context, a self-evolution skill that can acquire and install capabilities is especially dangerous because it could introduce unreviewed code, expand privileges, or create persistence paths through normal-looking workflow steps.

Missing User Warnings

High
Confidence
98% confidence
Finding
The listed shell commands operationalize search, installation, logging, and orchestration, but the documentation does not warn users that these actions may change the system or execute external code. This is risky because the skill frames these commands as routine steps in an evolution loop, making potentially dangerous actions appear normal and encouraging autonomous execution without review.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal