Back to skill

Security audit

zhongkui-skill

Security checks for vulnerabilities and agentic risk

Overview

This is a local skill-auditing helper whose file-reading behavior is disclosed and aligned with its security-review purpose.

Install this only if you want a local skill-review assistant. Give it explicit skill-directory paths, avoid pointing it at broad private folders, and treat its verdicts as heuristic because the permission boundaries and some advanced review claims are not tightly specified.

Vulnerability Patterns
  • 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
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (2)

Undeclared Tool Scope

Medium
Category
MCP Least Privilege
Confidence
91% confidence
Finding

The skill text describes capabilities that imply file reading, network access, and shell-oriented analysis steps, but it does not declare corresponding permissions. This creates a trust and review gap: operators may approve or run the skill without understanding the actual access it expects, increasing the chance of over-privileged execution or unsafe deployment.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
84% confidence
Finding

The activation phrases include broad everyday expressions such as '审一下' and '查一下这个skill', which can cause accidental invocation outside a clearly intended security-review context. Over-broad triggering is dangerous for a skill with analysis capabilities because it may activate unexpectedly, inspect unintended content, or interfere with normal conversations and workflows.

Content

No source excerpt is available for this finding.