google-agents-cli-observability

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

• Third-party content exposure detected (medium risk: 0.30). The feedback mechanism reference file describes a FastAPI endpoint (/feedback) that accepts user-submitted free text (text) and logs it into a pipeline monitored by BigQuery log sinks.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (medium risk: 0.30). The skill instructs the agent to fetch documentation and integration guides from adk.dev domains at runtime, which represents a third-party operational dependency.