aws-cost-operations_skill

Overview

This skill provides hands-on AWS cost optimization, monitoring, observability, and security assessment workflows with eight integrated MCP servers for billing analysis, pricing, metrics, traces, and audit. It helps estimate pre-deployment costs, analyze historical spend, set budgets and alarms, monitor application performance, and run security assessments against AWS best practices. The goal is to reduce waste, improve reliability, and maintain a secure, cost-efficient AWS environment.

How this skill works

The skill exposes MCP servers that connect to AWS Billing, Pricing, Cost Explorer, CloudWatch (metrics/logs), CloudWatch Application Signals, Managed Prometheus, CloudTrail, and a Well-Architected Security Assessment tool. Use the Pricing MCP to estimate TCO and compare regions before deployment, Billing and Cost Explorer MCPs to monitor and analyze actual spend, CloudWatch and Prometheus MCPs to observe performance and trigger alarms, and CloudTrail plus the Security Assessment MCP to audit and remediate security issues. Workflows combine pre-deployment estimation, continuous monitoring, cost analysis, and regular security auditing.

When to use it

• Estimate costs and compare regions before launching new services or features
• Detect and investigate unexpected AWS charges or cost anomalies
• Set budgets, alerts, and automated cost controls for teams and projects
• Monitor application health, latency, and resource utilization in production
• Audit API activity, track configuration changes, and investigate incidents
• Assess security posture against the AWS Well-Architected Security pillar

Best practices

• Estimate costs with the Pricing MCP before provisioning and include scaling scenarios
• Tag resources consistently for accurate cost allocation and reporting
• Right-size instances and use Savings Plans or Reserved Instances where appropriate
• Set actionable CloudWatch alarms with clear remediation runbooks to reduce noise
• Run CloudTrail audits and Well-Architected security assessments regularly
• Automate routine cleanups (idle resources, snapshots) and schedule non-production shutdowns

Example use cases

• Calculate monthly cost for a Lambda-heavy workload and compare across regions
• Detect a sudden spike in S3/EBS costs and drill into Cost Explorer reports to find the root cause
• Create CloudWatch dashboards and alarms for CPU, latency, and error budgets for an e-commerce app
• Query Prometheus metrics for Kubernetes pod memory pressure and create alerts for eviction risk
• Run a security assessment to find IAM privilege gaps and generate prioritized remediation tasks