ClawHub
Back to skill

Security audit

概念验证 验证总监工作台

Security checks for vulnerabilities and agentic risk

Overview

This is a disclosed Gate-review decision-template skill with no executable behavior or hidden data access.

Use this skill for concept-validation Gate reviews and related project status decisions. Treat its Go/No-Go, resource-allocation, and IP/safety conclusions as advisory unless reviewed by the responsible human stakeholders; the fallback contact email is disclosed and optional.

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
90% confidence
Finding
The trigger conditions are broad enough to activate on generic requests like asking about a project's current status, which can cause the agent to invoke this specialized decision-making workflow outside its intended context. That creates scope confusion, increases the chance of irrelevant role-played authority or unnecessary data collection, and may interfere with routing to more appropriate skills.

Static analysis

No suspicious patterns detected.