deep-insight

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

• Third-party content exposure detected (high risk: 1.00). The skill's workflow (SKILL.md — Bước 0 and 1.3 "Research bổ sung") explicitly accepts URLs/files and instructs the agent to use WebSearch/WebFetch/deep-research to read web sources, meaning it will fetch and interpret open/public third-party content as part of its analysis and decisions.