research-literature-review

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

• Third-party content exposure detected (medium risk: 0.30). SKILL.md describes that at runtime the workflow (Stage 3) uses LLM to “逐篇阅读并评分（标题与摘要）” from retrieved papers and OpenAlex results, which are outsider-authored text coming from bibliographic sources rather than first-party trusted content.