ai-research-assistant

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

• Third-party content exposure detected (high risk: 0.90). The skill's Literature Review workflow explicitly instructs the agent to "Search across arXiv, PubMed, IEEE, etc." and to summarize/analyze retrieved papers, so it fetches and ingests open/public third-party content (public websites) that the agent will read and use to drive decisions and next actions.