aws-advanced-patterns

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

• Third-party content exposure detected (medium risk: 0.30). The skill’s runtime artifacts (e.g., Step Functions/Lambda/EventBridge/SQS templates) would ingest outsider-provided event payloads/queue messages into workflow state (e.g., templates/step-functions-state-machine.json using ItemsPath / Payload mappings and handing $.detail into targets), and those payloads are free text authored by outsiders.