market-scout

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

• Third-party content exposure detected (high risk: 0.90). Yes — the skill's Deep-dive step (Step 4) explicitly requires "bottom-up validation" using public, user-generated sources like Reddit/forum threads, Amazon review patterns, Google Trends, and Upwork/Fiverr demand signals, which the agent would need to fetch and interpret as part of its workflow.