research-discover

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

• Third-party content exposure detected (high risk: 0.90). The skill's workflow explicitly expects web-backed literature retrieval and asks "能 websearch 还是用户提供链接" in SKILL.md and notes "web-backed literature mapping" in discovery-workflow.md, meaning the agent will fetch and interpret public third‑party webpages/papers (untrusted sources) as part of its required literature-verification and decision-making steps.