biorxiv-database

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

• Third-party content exposure detected (high risk: 0.90). This skill fetches and ingests content (titles, abstracts, full-text PDFs) directly from the public bioRxiv preprint server (e.g., https://www.biorxiv.org and DOI PDF URLs), which is untrusted/user-generated third-party content that the agent is expected to read and interpret as part of its workflow.