design-research

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

• Third-party content exposure detected (high risk: 1.00). This skill explicitly requires mining and analyzing public, user-generated App Store reviews and other external user feedback (see "App Store Review Analysis" and the scoping step listing App Store reviews, support tickets, and user emails), which the agent would ingest and interpret to drive product decisions, creating exposure to untrusted third-party instructions.