literature-review

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

• Third-party content exposure detected (high risk: 0.90). The skill's required workflow explicitly directs the agent to develop search queries and ingest/screen/extract content from public databases (e.g., Google Scholar, Semantic Scholar, arXiv, PubMed, ACM Digital Library) and to read and synthesize those third-party papers as part of its analysis, which exposes it to untrusted public content that could carry indirect prompt-injection risks.