alibabacloud-sre-toolkit

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

• Third-party content exposure detected (medium risk: 0.65). STAROps SSE streaming in scripts/starops_manager.py chat returns assistant text from the external STAROps service, which is then incorporated into the workflow’s diagnostic/report generation and thus into the agent’s LLM-visible context; this is indirect prompt injection exposure from an outsider-controlled endpoint.