auto-updater

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The skill explicitly instructs the agent to display full diffs (including hooks/hooks.json and .mcp.json which can contain credentials or commands run with your credentials), which would require outputting any secret values found verbatim, creating an exfiltration risk.

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

• Third-party content exposure detected (high risk: 0.85). SKILL.md 第2步/2.5步要求“展示完整差异”并在扫描中读取“新 SKILL.md”（来自上游/源注册表拉取的新版本内容），属于公共/外部来源的自由文本被注入到LLM上下文的路径。