ClawHub
Back to skill

Security audit

测试领域建模

Security checks for vulnerabilities and agentic risk

Overview

This skill is a documentation-oriented QA modeling helper with read-only workspace access and no hidden execution, persistence, or data-export behavior.

Install only if you want an agent to read relevant project requirements or scenario documents and turn them into QA domain models. Be aware that the trigger terms are broad, so invoke it deliberately for QA modeling work rather than generic design conversations.

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
89% confidence
Finding
The trigger conditions are broad natural-language phrases such as “建模”, “数据流”, and general descriptions like needing to understand complex business processes, which can cause the skill to activate outside its intended scope. In an agent system, over-broad invocation can lead to workflow confusion, irrelevant outputs, and accidental exposure of contextual project information to a skill that was not deliberately selected.

Static analysis

No suspicious patterns detected.