gemini-masterpiece

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

• Third-party content exposure detected (high risk: 1.00). The skill explicitly requires "Mandatory Google Search grounding for active research" and instructs use of "native URL processing for real-time market research" in SKILL.md, which means the agent will fetch and interpret open/public web content that could contain untrusted, user-generated instructions affecting its actions.