alphaxiv

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

• Third-party content exposure detected (high risk: 0.80). This skill explicitly fetches and ingests machine-readable paper overviews from AlphaXiv's public API (GET https://alphaxiv.org/api/papers/{versionId}/overview) and can fall back to fetching arXiv pages (https://arxiv.org/abs/{id}), and those third‑party summaries are read and used to populate notes, suggest tags, and drive linking actions—so untrusted external content could indirectly influence agent behavior.