ClawHub
Back to skill

Security audit

LiuZewei Hello Demo

Security checks for vulnerabilities and agentic risk

Overview

This is a small greeting demo that only prints local messages and shows no evidence of credential access, persistence, networking at runtime, or destructive behavior.

This appears safe to install as a demo greeting skill. Review the small Python script if you want extra assurance, and treat the GitHub update URL as informational unless your installer separately asks to enable automatic updates.

SkillSpector

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
86% confidence
Finding
The example shows the skill being triggered by a very generic user request ('Say hello to Alice') with no visible namespace, confirmation step, or tighter intent boundaries. In an agent environment, overly broad invocation patterns can cause accidental tool execution on common conversational input, increasing the chance of unintended actions and creating an unsafe precedent for more capable skills.

Static analysis

No suspicious patterns detected.