lov-finder-action

CRITICAL E006: Malicious code pattern detected in skill scripts.

• Malicious code pattern detected (high risk: 1.00). The skill includes explicit patterns that grant broad filesystem entitlements, monitor the entire filesystem, and implement a helper-app + clipboard chain that executes arbitrary commands — enabling remote code execution and data access/exfiltration if abused.

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (medium risk: 0.30). 运行路径是：根据用户消息触发“模式判断/Step 1 收集需求”，再把用户提供的“动作名称/触发描述/核心命令/文件类型”等自由文本写入生成的 Automator workflow 或 Finder Sync Extension 代码与脚本，因此自由文本会被 LLM 直接消费而无需先选择特定外部项目。

MEDIUM W013: Attempt to modify system services in skill instructions.

• Attempt to modify system services in skill instructions detected (high risk: 0.80). The skill instructs creating and installing apps under /Applications (mkdir/cp into /Applications, writing executable helper scripts) and recommends entitlements allowing absolute-path read-write, which modifies system-wide locations and may require or encourage elevated privileges and can compromise the host.