Optimize AWS Cloud Spending

This skill analyzes AWS spending patterns, identifies waste, and produces actionable cost reduction strategies. Its cost analysis capability parses AWS Cost Explorer data for trends and anomalies, breaks costs down by service/region/resource tag, and flags month-over-month spending increases. Its resource optimization capability detects idle EC2 instances via low CPU utilization, finds unattached EBS volumes and old snapshots, identifies unused Elastic IPs, locates underutilized RDS instances, and finds old S3 objects eligible for lifecycle policies. Its savings recommendations cover Reserved Instance/Savings Plans opportunities, CloudWatch-metrics-based instance rightsizing, resources sitting in expensive regions, and calculated potential savings for specific actions. It provides ready-to-run AWS CLI commands for each of these: aws ce get-cost-and-usage for monthly-by-service or daily-in-month cost breakdowns; aws ec2 describe-volumes/describe-addresses for unattached EBS volumes and unused Elastic IPs; aws cloudwatch get-metric-statistics against EC2 or RDS instances for idle/utilization detection; aws ec2 describe-snapshots filtered to snapshots older than 90 days; and aws ec2 describe-instances for a rightsizing inventory of instance types. The optimization workflow moves through four stages: baseline assessment (3-6 months of cost data, top 5 spending services, growth rate), quick wins (delete unattached volumes, release unused Elastic IPs, stop/terminate idle instances, delete old snapshots), strategic optimization (Reserved Instance coverage, instance-type-vs-workload review, S3 lifecycle policies, Spot instances for non-critical workloads), and ongoing monitoring (AWS Budgets with alerts, Cost Anomaly Detection, resource tagging, monthly review meetings) - backed by a ten-item checklist covering Cost Explorer, budgets, unused-resource cleanup, Reserved Instances, S3 Intelligent-Tiering, data transfer costs, Lambda memory sizing, and CloudWatch Logs retention. It also integrates with Kiro CLI (kiro-cli chat "Use aws-cost-optimizer to analyze my spending") for conversational invocation.

Use this skill when you need to analyze AWS spending, identify cost optimization opportunities, or reduce cloud waste - from a quick unused-resource sweep to a full Reserved Instance and rightsizing strategy. It is explicit about risk tiering: read-only analysis is low risk, deletion actions are medium risk and must always be verified before executing, and production rightsizing changes are high risk and should be tested in dev/staging first, with backups maintained before any deletion and a --dry-run flag used where available.

Input is AWS account access (Cost Explorer, EC2, RDS, CloudWatch, S3 permissions) and a target time window for analysis. Output is cost breakdowns and anomaly reports, a list of specific waste (idle instances, unattached volumes, unused IPs, old snapshots) with calculated savings, and prioritized recommendations (Reserved Instances, rightsizing, lifecycle policies) following the baseline-quick wins-strategic-monitoring workflow.

Built on the AWS CLI (Cost Explorer ce, ec2, cloudwatch commands), AWS Cost Anomaly Detection, AWS Budgets, AWS Trusted Advisor, and Kiro CLI for conversational invocation.

aws ce get-cost-and-usage \
  --time-period Start=$(date -d '30 days ago' +%Y-%m-%d),End=$(date +%Y-%m-%d) \
  --granularity MONTHLY \
  --metrics BlendedCost \
  --group-by Type=DIMENSION,Key=SERVICE

Cloud engineers, FinOps practitioners, or teams managing AWS costs who need concrete CLI commands and a structured workflow to find waste and cut spend, with explicit risk guardrails for anything beyond read-only analysis.