mle-agent-guide

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

• Third-party content exposure detected (high risk: 0.90). The skill's "arXiv Integration" section explicitly describes monitoring, fetching, and extracting methods from arXiv papers (public, user-submitted preprints), which the agent is expected to read and act on to generate implementation plans and drive experiments, so untrusted third‑party content could influence its actions.