research-workflow

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

• Third-party content exposure detected (high risk: 0.90). The skill's mandatory Gathering stage ("阶段2: 收集 (Gathering)") requires collecting "相关文档" and in examples explicitly "收集各框架文档、社区讨论、性能测试", which implies ingesting open/public user-generated content (community discussions) that the agent will read and use to drive analysis and recommendations, exposing it to potential indirect prompt injection.