bedrock-automated-reasoning

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 1.00). The skill loads policy documents from S3 at runtime (e.g., s3://my-policy-documents/healthcare/hipaa-clinical-protocols.pdf and s3://my-policy-documents/healthcare/hipaa-clinical-protocols-v2.pdf) to create/update automated-reasoning policies whose extracted rules directly control validation/response behavior, so this external content can change agent instructions at runtime.