ai-system-testing

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

• Third-party content exposure detected (high risk: 0.90). SKILL.md explicitly requires testing tool calls that invoke web_search (Tool Call Validation) and RAG grounding verification plus extracting/HEAD‑checking URLs (Hallucination Risk Assessment), which involve fetching and interpreting external, untrusted third‑party documents that can influence tool selection and subsequent decisions.