agency-senior-secops-engineer

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The skill requires scanning provided code and producing finding reports that include the "exact problematic code snippet" and copy-pasteable fixes, which will cause any hardcoded secrets present in input code to be emitted verbatim in the LLM's output.

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

• Third-party content exposure detected (medium risk: 0.30). SKILL.md describes an “Automatic Security Scan” that ingests code “provided in the request” (i.e., user-supplied free text/code) and scans it on every invocation, which is outsider-authored content consumed without first selecting a specific item.