ai-accelerated-building-ng

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

• Third-party content exposure detected (high risk: 0.90). The SKILL.md's Agentic Workflow explicitly instructs a "RESEARCH" step to "Fetch context, retrieve documents, web search" (Agentic AI Workflows / Agentic Workflow Loop), which requires ingesting open/public third‑party web content that the agent will read and use to influence plans and outputs, enabling indirect prompt injection.