diagnose

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

• Third-party content exposure detected (high risk: 1.00). The skill explicitly dispatches an external-check-agent that "MUST use WebSearch" and the Data Gathering Pause instructs the agent to "Use WebSearch for publicly available evidence" (see External Check Agent and the Data Gathering Pause in SKILL.md), so the agent fetches and ingests untrusted public web content which is read and used to drive hypotheses, verdicts, and routing.