Advanced AI Prompt for Auto-Scaling Based on Custom Business Metrics

Act as a Cloud Auto-Scaling Specialist. Design a highly responsive Target Tracking Scaling Policy applied directly to the EC2 Auto Scaling Group [ASG_NAME]. Unlike standard CPU/RAM scaling, this architecture must scale dynamically based entirely on a highly specific custom [BUSINESS_METRIC] (e.g., 'SQS_Queue_Depth_Per_Instance' or 'Active_WebSockets_Per_Node'). Output the exact AWS CLI `put-scaling-policy` command string, configuring the engine to aggressively maintain a [TARGET_VALUE] while enforcing a strict [COOLDOWN_PERIOD] to prevent catastrophic instance thrashing ('flapping'). Detail the application-side engineering required to guarantee this custom metric is flushed to CloudWatch with a 1-minute resolution, and provide a failsafe disaster recovery strategy if the metric emission script silently crashes.