ericyanpek/agentcore-private-registry-blueprint / skill-package/src/aws_cost_anomaly_triage/skill_files
aws-cost-anomaly-triage
VerifiedUse when the user reports an unexpected AWS cost spike, an alert from AWS Cost Anomaly Detection, or asks "why did my AWS bill go up". Walks through a structured triage SOP using Cost Explorer, CUR, and service-specific signals before recommending remediation.
AWS Cost Anomaly Triage
This skill encodes the internal SOP our FinOps team uses when an AWS cost anomaly is raised. Apply it whenever you see a cost spike alert, an unfamiliar line item on the bill, or a customer-reported overage.
When to invoke
- An anomaly is reported by AWS Cost Anomaly Detection (email, Slack, EventBridge)
- Month-over-month spend on any service rises by more than the threshold in
resources/thresholds.md - A customer or team asks "why is my AWS bill higher than expected"
Triage workflow
Follow these steps in order. Do not skip ahead even if the cause seems obvious — early steps surface confounding factors that shape later recommendations.
Step 1 — Bound the anomaly
- Identify the time window (start, end, ongoing?). Use Cost Explorer
GetCostAndUsagegrouped bySERVICEover the past 30 days at daily granularity. - Identify the scope: account, region, service, usage type, linked
account, or tag dimension. Always confirm the anomaly persists when
you remove plausible confounders (RI/SP amortization, credits,
refunds — see
resources/confounders.md). - Quantify: absolute $ delta and % delta. If <$50 or <5%, downgrade to monitoring rather than active triage.
Step 2 — Classify
Match against the patterns in resources/anomaly-patterns.md. The
common ones in order of frequency:
- Compute scale-up (EC2/Fargate/Lambda)
- Data transfer (especially cross-AZ and NAT GW egress)
- Storage growth (S3 standard creep, EBS snapshots)
- New service activation (Bedrock, OpenSearch, Aurora upgrade)
- Tagging gap surfacing previously hidden cost
- Credit / discount expiration
Step 3 — Drill down
Use service-specific dashboards listed in
resources/per-service-drilldown.md. Always check:
- CloudTrail for unusual
Run*/Create*API spikes around the anomaly start time - Cost and Usage Report (CUR) at hourly granularity if Cost Explorer daily isn't enough
- Resource tags to attribute to a team or workload
Step 4 — Decide remediation
Map the classified cause to one of the responses in
resources/remediation-playbook.md. Never propose:
- Disabling a security service (GuardDuty, Macie, Detective) without the security team's written sign-off
- Deleting CloudTrail logs or VPC Flow Logs to save storage
- Removing tags to "simplify"
Step 5 — Document
Write up findings using the template in resources/postmortem-template.md.
File it in the FinOps tracker even if cause was benign — the patterns
compound across incidents.
Resources
- Confounders to rule out
- Anomaly patterns library
- Per-service drilldown links
- Remediation playbook
- Postmortem template
- Internal thresholds
What this skill is NOT
- Not a substitute for Cost Optimization Hub or Trusted Advisor recommendations — those are starting points, not conclusions.
- Not for capacity planning or pricing-model selection (RI/SP). Use the
aws-savings-plan-modelingskill for those.
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| writing-skillsobra/superpowersUse when creating new skills, editing existing skills, or verifying skills work before deployment | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| xlsxanthropics/skillsUse this skill any time a spreadsheet file is the primary input or output. This means any task where the user wants to: open, read, edit, or fix an existing .xlsx, .xlsm, .xltx, .csv, or .tsv file (e.g., adding columns, computing formulas, formatting, charting, cleaning messy data); create a new spreadsheet from scratch or from other data sources; or convert between tabular file formats. Trigger especially when the user references a spreadsheet file by name or path — even casually (like "the xlsx in my downloads") — and wants something done to it or produced from it. Also trigger for cleaning or restructuring messy tabular data files (malformed rows, misplaced headers, junk data) into proper spreadsheets. The deliverable must be a spreadsheet file. Do NOT trigger when the primary deliverable is a Word document, HTML report, standalone Python script, database pipeline, or Google Sheets API integration, even if tabular data is involved. | Stars: 198k | Updated | Claude CodeCodex | Verified | Install |
| shipping-and-launchaddyosmani/agent-skillsPrepares production launches. Use when preparing to deploy to production, or when asking what needs to be in place before shipping. Use when you need a pre-launch checklist, when setting up monitoring, when planning a staged rollout, or when you need a rollback strategy. | Stars: 119k | Updated | Claude CodeCodex | Verified | Install |
| childhoodthedotmack/claude-memBabysit a shipped release after merge, publication, or deployment. Verify published packages, production behavior, installed plugins, and late CI or user reports until the observation window is clean; use babysit for PR readiness before release. | Stars: 118k | Updated | Claude CodeCodex | Verified | Install |