olares-cluster

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

• Third-party content exposure detected (high risk: 0.75). Outsider free text can enter the LLM context via cluster pod logs -f / cluster pod get -w / cluster workload rollout-status -w / cluster application status -w, because the agent ingests server-returned log/event text (which may include arbitrary user-authored content from other tenants/apps) into the runtime prompt.