ClawHub
Back to skill

Security audit

GitHub开发规范工具

Security checks for vulnerabilities and agentic risk

Overview

The skill appears to be a broad GitHub development helper, but it requests shell and file-modification authority without enough concrete limits or user-control details.

Review this skill before installing. It is not clearly malicious, but users should only invoke it on repositories they trust, confirm every file change and shell command, and avoid providing API keys or credentials unless the exact operation and destination are clear.

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
93% confidence
Finding
The skill advertises high-risk command execution capability via the `exec` tool while claiming that only whitelisted commands are executed, but the file contains no actual whitelist, policy, or enforcement mechanism. This can mislead users and downstream agents into trusting shell execution in broad development scenarios, increasing the chance of unsafe command use with user-controlled inputs.

Vague Triggers

Medium
Confidence
87% confidence
Finding
The trigger and usage description are so broad that the skill can match routine coding, debugging, deployment, and development-assistance requests. In a skill with `read`, `write`, and `exec`, overly broad activation increases the likelihood that the agent invokes a powerful workflow in contexts where the user did not specifically intend shell-capable automation.

Vague Triggers

Medium
Confidence
84% confidence
Finding
The scenario and capability language uses vague phrases such as broad 'GitHub' handling and generic 'processing' outcomes, without clearly defining what operations are in scope. This ambiguity is dangerous because it expands effective authority for a skill that includes command execution and file modification, making accidental or excessive use more likely.

Static analysis

No suspicious patterns detected.