xhs-skill

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.90). The prompt explicitly requires the agent to run a local QR-decoding command and "直接发给用户" the decoded QR text (and ASCII QR), which forces the LLM to output a likely one-time login token/credential verbatim (a sensitive secret), even though other secrets like cookies are constrained to local files.