ClawHub
Back to skill

Security audit

代理副驾驶

Security checks for vulnerabilities and agentic risk

Overview

This is a markdown guidance skill for agent and prompt engineering, with no artifact-backed evidence of hidden execution, data collection, persistence, or destructive behavior.

Before installing, be aware that the skill may appear in broad prompt-engineering, tool-selection, or task-decomposition contexts. Its exec declaration should be treated as optional authority for validation tasks, and users should review any proposed command before allowing execution.

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
88% confidence
Finding
The trigger keywords include broad conceptual terms such as "prompt工程", "工具选择", and "task decomposition" without scope constraints, so the skill may activate in many unrelated conversations. In a skill with powerful capabilities and declared `exec` access, unintended invocation increases the chance of unnecessary tool suggestions or workflow steering in contexts the user did not intend.

Static analysis

No suspicious patterns detected.