amazon-market-trend-scanner

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

• Third-party content exposure detected (high risk: 0.80). High: the required workflow reads external, outsider-authored free text from ZooData API review content (e.g., product review titles/bodies) and injects it into the agent/LLM context via the “prompt-as-data” path that renders review-tag-prompt using {text} (caller’s LLM consumes that prompt).