sci-journal-submission-expert

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

• Third-party content exposure detected (high risk: 0.90). The skill explicitly asks users to provide or paste target journal "Author Guidelines" links or original excerpts ("需要期刊细则时请用户粘贴 Author Guidelines 链接或原文片段以便逐项核对") and states it will use official, publicly available guidelines as anchors, meaning the agent will ingest and act on untrusted third‑party web content as part of its workflow.