evidence-first-debugging

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.80). The prompt requires including exact commands and verification commands (e.g., the "command: " field and passing verification evidence), which would force the model to emit any embedded credentials or tokens verbatim if those appear in observed commands, and it gives no guidance to mask or use environment-based auth.