workflow-agent

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

• Third-party content exposure detected (high risk: 0.90). The skill explicitly instructs the Explorer to perform WebSearch/WebFetch (see reference/explorer.md "4단계: 탐색 실행") and the planner's <reference> mapping permits workers to Read/WebFetch arbitrary URLs, so untrusted public web content is ingested and can materially influence agent decisions and actions.