amazon-review-intelligence-extractor

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

• Third-party content exposure detected (high risk: 0.95). In the required “Insufficient Data Fallback” workflow, the agent fetches outsider-authored Amazon review bodies at runtime (via zoodata.py reviews-raw calling realtime/reviews) and then feeds that free text into its own LLM context for Map/Reduce tagging (review-tag-prompt / review-reduce-prompt using Review Text: """{text}""").