amazon-listing-audit-pro

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

• Third-party content exposure detected (medium risk: 0.30). Within the listing-audit workflow, the agent retrieves raw product reviews via the ZooData API (reviews-raw / realtime/reviews-style data) and then renders per-review prompts by embedding the fetched review JSON/text into the LLM prompt template (prompt-as-data), so outsider-authored review text can be consumed at runtime.