adversarial-review

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

• Third-party content exposure detected (high risk: 1.00). The skill's mandatory Stage 1 ("Reality vs Claims") explicitly requires using web_search and web_fetch "min 3x diverse" (including library docs, forum complaints, benchmarks, post-mortems and other public web sources), so the agent will fetch and interpret untrusted, user-generated third‑party content that can influence subsequent decisions and actions.