ClawHub
Back to skill

Security audit

audit

Security checks for vulnerabilities and agentic risk

Overview

The skill appears to be a broad security-audit helper, but it requests read, write, and command-execution authority without enough scoping or enforceable safety controls.

Review this skill carefully before installing. It is not clearly malicious, but only use it in a controlled workspace where you are comfortable allowing an agent to read and write project files and run commands. Avoid using it on sensitive repositories or production environments unless you add your own command restrictions and require explicit approval for changes or deployments.

SkillSpector

Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (3)

Intent-Code Divergence

Medium
Confidence
94% confidence
Finding
The skill claims that only whitelisted commands are executed, yet the manifest exposes a general `exec` capability and the Markdown provides no actual whitelist, policy, or enforcement mechanism. This creates a misleading safety guarantee that could cause users or downstream agents to trust command execution paths that are in fact unconstrained.

Vague Triggers

High
Confidence
88% confidence
Finding
The invocation description is overly broad, covering security auditing, code generation, debugging, deployment, and general automation contexts without clear scope boundaries. In a skill that has `read`, `write`, and especially `exec`, ambiguous scope increases the chance an agent will perform impactful operations in unintended contexts or under weak user authorization.

Missing User Warnings

Medium
Confidence
92% confidence
Finding
The skill omits a prominent warning that it may read files, write files, and execute system commands. Because the skill declares powerful capabilities, lack of an upfront warning can mislead users into invoking it without understanding the operational and security impact, increasing the risk of unsafe execution or unintended system changes.

Static analysis

No suspicious patterns detected.