academic-deep-research

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 (Cycle Architecture and Cycle Execution Protocol) explicitly directs searching public third-party sources such as "Google Scholar, Semantic Scholar, PubMed, IEEE, SSRN, arXiv, bioRxiv" and producing an annotated bibliography/evidence matrix based on those findings, meaning the agent will ingest and act on untrusted, user-generated open-web content.