research-patterns

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

• Third-party content exposure detected (high risk: 0.90). The skill's "External Research" step explicitly instructs the agent to "Use WebSearch" and search external documentation and to store findings with a source URL (e.g., metadata: source: ""), meaning the agent will fetch and read open/public web content that can include untrusted third-party user-generated material.