Back to skill

Security audit

Coding Rules

Security checks for vulnerabilities and agentic risk

Overview

This is a project coding-standards skill for Vue3 and TypeScript work, with no executable payload, credential handling, persistence, or hidden high-impact behavior.

Install this if you want agents to follow these Vue3 project conventions. Be aware that it may influence most Vue3 + TypeScript coding tasks in the repository and the instructions are written in Chinese, so teams should ensure that matches their workflow.

Vulnerability Patterns
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs
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 (2)

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
84% confidence
Finding

The skill description is written to operate in Chinese and does not offer a user-language choice or document a hard locale requirement. In multi-user or mixed-language environments, this can degrade usability, cause misunderstanding of generated guidance, and create a mismatch between user intent and agent behavior.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The trigger condition is explicitly broad: it activates whenever a user requests generation, modification, or review of Vue3 + TypeScript subproject code, even if the user did not ask for this specific skill. In an agent system, overly broad auto-activation can cause unintended instruction injection, where project-specific rules override user intent or leak into unrelated coding tasks within a monorepo.

Content

No source excerpt is available for this finding.