operational-readiness

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.80). The prompt instructs the agent to "grep for hardcoded secrets" and scan infra/code for secrets (e.g., .tf/.ssm references), which requires reading potential secret strings and does not instruct redaction or forbids echoing them, creating a significant exfiltration risk.