alterlab-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 bibliography_agent explicitly performs systematic searches and ingests results from external databases and "other sources" (agents/bibliography_agent.md Step 2: Execute Search), and the monitoring_agent configures alerts against public platforms like Google Scholar, PubMed, and Retraction Watch (agents/monitoring_agent.md), meaning the agents read and act on untrusted, third-party web content as part of their required workflow.