product-reviews-ratings

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

• Third-party content exposure detected (high risk: 0.80). The skill explicitly accepts and displays untrusted, user-generated product reviews (see SKILL.md and the submission/display endpoints like POST /api/reviews, review-api.ts, and review-submission.ts), and those review bodies are read and used to make moderation and publishing decisions (auto-approve, spam/profanity checks, updateProductRatingSummary), so third-party content can materially influence agent actions.