ClawHub
Back to skill

Security audit

Verify Code Change

Security checks across malware telemetry and agentic risk

Overview

This skill is a straightforward test-runner helper that executes user-provided verification commands, with no hidden persistence or data exfiltration behavior found.

Install only if you are comfortable with a helper that runs test commands in your current project. Use it with trusted repositories and explicit test commands, because project test scripts can run arbitrary code even though this skill itself does not add hidden behavior.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Findings (4)

Lp3

Medium
Category
MCP Least Privilege
Confidence
95% confidence
Finding
The skill advertises shell-command execution capability through its examples and functionality, but does not declare permissions or constraints for that capability. This is dangerous because an agent may execute arbitrary user-supplied commands in the project root, enabling command injection, destructive filesystem actions, credential exposure, or execution of malicious package scripts without any explicit permission boundary.

Vague Triggers

Medium
Confidence
89% confidence
Finding
The activation text, 'When you need to verify functionality after code changes,' is broad enough to trigger in many ordinary development contexts without clear limits. In a skill that can execute shell commands, vague activation criteria increase the chance the agent invokes it automatically in situations where command execution was not explicitly intended or sufficiently reviewed.

Vague Triggers

Medium
Confidence
94% confidence
Finding
The usage text states that providing a test command causes the script to 'automatically execute and check results,' which normalizes direct execution of general user input. Because the accepted parameter is a free-form command string, this can lead to arbitrary shell execution if a user, prompt, or upstream tool supplies unsafe input.

Autonomous Decision Making

Medium
Category
Excessive Agency
Content
## Usage

Provide test command, script will automatically execute and check results.

## Parameters
Confidence
91% confidence
Finding
automatically execute

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.