literature-review

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

• Third-party content exposure detected (high risk: 0.90). The skill explicitly instructs the agent to search and ingest content from open public sources (e.g., via gget for PubMed and bioRxiv, direct API/WebFetch for arXiv, Semantic Scholar and Google Scholar) and requires reading abstracts and full texts and extracting/synthesizing their content, so it clearly consumes untrusted third-party material as part of its workflow.