make-scenario-building

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

• Third-party content exposure detected (high risk: 0.90). The skill explicitly instructs agents to fetch and interpret live external, user-generated content (e.g., emails, Zendesk tickets, Slack/Discord messages, webhooks, HTTP/Weather APIs) as runtime inputs that the AI Agent will read and use to decide which tools/actions to call (see ai-agents.md examples and SKILL.md Phase 2 / Quick Patterns describing tool attachment and runtime tool-calling).