ClawHub
Back to skill

Security audit

Security Auditor

Security checks for vulnerabilities and agentic risk

Overview

This is a general security-auditing skill whose read and command-execution capabilities are disclosed and fit its stated purpose.

Before installing, be prepared to give it access only to the codebase you want reviewed and to approve any command execution deliberately, especially if the repository contains secrets or production configuration.

SkillSpector

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
92% confidence
Finding
The skill’s invocation guidance is extremely broad, covering general security review, auth implementation, dependency auditing, compliance, encryption, and automation use cases without clear activation boundaries or prerequisites. Overly broad routing increases the chance the agent is invoked in inappropriate contexts and may overreach into sensitive repositories or system actions without sufficient user intent validation.

Vague Triggers

Low
Confidence
84% confidence
Finding
The startup instructions tell users to call the skill generically with necessary parameters, but do not define precise activation cues or safe operating boundaries. This can cause accidental or excessive invocation, especially in agent environments where generic descriptions influence tool selection automatically.

Missing User Warnings

Medium
Confidence
94% confidence
Finding
The skill advertises file handling, API integration, and command execution, but it does not place a prominent upfront warning that these capabilities may affect the local system or external services. In an agentic environment, normalizing execute-capable behavior without conspicuous consent language can lead to unintended command execution, filesystem modification, or sensitive data exposure.

Static analysis

No suspicious patterns detected.