fund-output-template

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

• Third-party content exposure detected (high risk: 1.00). The skill explicitly requires using web-based sources (see the JSON example "sources" array with { "type": "web", "url": "https://..." } and the checklist item "웹검색 근거 출처 테이블 (필수)"), so the agent is expected to fetch and interpret public web content as part of its workflow and those external sources can materially influence portfolio decisions like SAFE_ASSET_DECISION.