competition-ios-runtime

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The skill explicitly instructs recording Keychain items, plaintext inputs/headers, local tokens and signed strings and replaying them, which requires capturing and outputting secret values verbatim.

CRITICAL E006: Malicious code pattern detected in skill scripts.

• Malicious code pattern detected (high risk: 0.90). 该技能明确指导绕过证书钉扎、hook Keychain 与请求签名边界、提取本地令牌并重放受保护请求，直接支持凭据窃取与未经授权的数据/会话重放，具有高度滥用风险。

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

• Third-party content exposure detected (high risk: 0.75). 该技能的运行流程要求对“Info.plist/Keychain/本地存储/网络请求构建/信任评估/钩子输出”等进行动态读取与记录；其中“钩子日志/返回值/请求头/请求体/本地存储内容”会以可读文本形式进入LLM上下文，而这些内容来源于被分析的IPA（非操作用户自有），属于外部文档/工件内容经运行时提取后的自由文本暴露路径。