plan-happy-path

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The skill repeatedly requires capturing and echoing the user's original request verbatim (passing it to child agents and writing it into a saved plan file), so any secrets present in the user's request would be included in LLM outputs and persisted, creating a high exfiltration risk.