zettaranc-perspective

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The skill explicitly asks the user to paste their Tushare Token and then embeds that token verbatim into python command strings and a write_env_file call, which requires the LLM to handle/output the secret directly (high exfiltration risk).

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

• Third-party content exposure detected (medium risk: 0.30). 运行时会在用户请求个股分析/选股/诊断/复盘等路径下读取用户自然语言输入（如股票代码、交易记录文字、口语化交易描述），并与后续工具返回的结构化数据一起被 LLM 用于生成回复；例如 SKILL.md 明确了契约输入包含“用户问题/意图”“交易记录/交易描述”等用户提供文本。