arxivterminal

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

• Third-party content exposure detected (high risk: 0.70). The skill's SKILL.md and references/arxivterminal-fetch.md explicitly fetch papers from the public arXiv API and references/arxivterminal-show.md/ search docs show the agent reads and displays paper titles/abstracts from that public, user‑provided content, so untrusted third‑party text is ingested and can influence search/results and subsequent actions.