paper-search

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

• Third-party content exposure detected (high risk: 0.80). The skill explicitly searches, downloads, and reads papers from open public sources such as arXiv, PubMed, Semantic Scholar, CrossRef, and others (see "Sources" and the "read <paper_id>" workflow in SKILL.md), so the agent ingests untrusted third-party content that could contain adversarial instructions affecting its actions.