From argos
Bulut maliyet review — right-sizing, idle detection, reserved/spot, egress, DB/storage tier, image retention, tag policy, FinOps maturity (Crawl/Walk/Run), anomaly alert.
npx claudepluginhub resultakak/argos --plugin argosThis skill uses the workspace's default tool permissions.
`agents/shared/severity-rubric.md` ve `agents/shared/escalation-matrix.md` default-load
Mandates invoking relevant skills via tools before any response in coding sessions. Covers access, priorities, and adaptations for Claude Code, Copilot CLI, Gemini CLI.
Share bugs, ideas, or general feedback.
agents/shared/severity-rubric.md ve agents/shared/escalation-matrix.md default-load
sayılır (agents/coordination.md §11). Bu skill'in çıktısı Critical / High / Medium /
Low + kanıt formatında olmak zorunda — spekülatif Critical yasak. Sahiplik dışı bulgu
ilgili agent'a delege; karar yetkisi eşiği aşılırsa kullanıcı onayı zorunlu.
| Veri kaynağı | İçerik |
|---|---|
| AWS Cost Explorer / GCP BigQuery billing / Azure Cost Mgmt | aylık + 90 gün trend |
| Cost & Usage Report (CUR) | line-item detay (resource bazlı) |
| Trusted Advisor / Compute Optimizer / Recommender API | otomatik right-sizing önerisi |
| Tag report | untagged % + tag coverage |
| Reserved/Savings Plan utilization | mevcut commit'ten yararlanma |
| Forecast | sonraki ay tahmini |
Snapshot:
Last 30 days: $48,200
- EC2 + EBS: $19,400 (40%)
- RDS: $11,500 (24%)
- S3: $4,800 (10%)
- Egress: $5,200 (11%)
- ALB + NAT GW: $3,100 (6%)
- Other: $4,200 (9%)
Forecast next 30: $52,000 (+8%)
Untagged: 12%
RI/SP coverage: 28% (baseline 60%+ hedef)
RI utilization: 91% (good)
| Boyut | Crawl | Walk | Run | Mevcut |
|---|---|---|---|---|
| Görünürlük | Aylık PDF | Cost Explorer | Real-time + ML | Walk |
| Sahiplenme | Tartışmalı | Tag policy | Otomasyon | Crawl-Walk |
| Optimizasyon | Manuel | RI plan | Continuous | Crawl |
| Anomaly | "Fatura geldi" | Alert | Forecast | Crawl |
| Sahiplik | Finance | Eng + Finance | FinOps team | Eng + Finance |
Hedef: 6 ayda Walk → Run.
# AWS untagged resource
aws resourcegroupstaggingapi get-resources --tag-filters Key=Owner,Values= \
--query 'ResourceTagMappingList[].ResourceARN' \
| wc -l # tagged
aws resourcegroupstaggingapi get-resources \
--query 'ResourceTagMappingList[?!(Tags[?Key==`Owner`])].ResourceARN' \
| wc -l # untagged
Hedef:
common_tags.Compute:
RDS:
Storage:
# AWS RI utilization (last 30d)
aws ce get-reservation-utilization \
--time-period Start=2026-04-10,End=2026-05-10
Karar:
Uygun workload (Crawl → Walk geçişi):
Uygun değil: prod web tier, primary DB, latency-sensitive.
# AWS data transfer cost (last 30d)
aws ce get-cost-and-usage \
--filter file://egress-filter.json \
--time-period Start=2026-04-10,End=2026-05-10 \
--metrics BlendedCost \
--granularity DAILY
Mitigation:
| Resource | Detection | Aksiyon |
|---|---|---|
| Unattached EBS | aws ec2 describe-volumes --filter Name=status,Values=available | snapshot + sil |
| Unused EIP | aws ec2 describe-addresses --filter Name=association.allocation-id,Values=null | release |
| Idle ALB (0 traffic 7d) | CloudWatch RequestCount 0 | sil |
| Unused RI / SP | utilization < %50 | exchange / sat |
| Old snapshot (> 90 gün) | tag age | lifecycle policy |
| Stopped EC2 (> 30 gün) | terminate veya sil | terminate |
| Untagged dev resource | resourcegroupstaggingapi | sahip bul, tag veya sil |
Auto-suspend dev/staging:
aws-nuke Friday cleanup.# AWS ECR untagged image lifecycle
aws ecr put-lifecycle-policy --repository-name api-svc \
--lifecycle-policy-text '{"rules":[
{"rulePriority":1,"description":"untagged > 7d","selection":
{"tagStatus":"untagged","countType":"sinceImagePushed","countUnit":"days","countNumber":7},
"action":{"type":"expire"}},
{"rulePriority":2,"description":"PR < 30d","selection":
{"tagStatus":"tagged","tagPrefixList":["pr-"],"countType":"sinceImagePushed","countUnit":"days","countNumber":30},
"action":{"type":"expire"}}
]}'
# AWS Budget per-service
aws budgets create-budget \
--account-id $ACCOUNT \
--budget '{"BudgetName":"api-svc-monthly","BudgetLimit":{"Amount":"3000","Unit":"USD"},
"TimeUnit":"MONTHLY","BudgetType":"COST",
"CostFilters":{"TagKeyValue":["user:Service$api-svc"]}}'
# Cost Anomaly Detection
aws ce create-anomaly-monitor --anomaly-monitor file://monitor.json
Slack/PagerDuty integration:
Quick win (≤ 30 gün):
Medium (≤ 90 gün):
Strategic (≤ 1 yıl):
aws_instance console click (IaC dışı).latest image tag prod (registry kontrolsüz).User: /cost-review
Agent (iac-engineer):
1. Cost Explorer son 30d: $48,200; trend +8%/ay.
2. Tag coverage: %88 (Untagged 12% — Critical).
3. Right-sizing:
- 4 EC2 t3.large CPU avg %18 → t3.medium (-$200/ay).
- RDS db.r5.xlarge CPU %22 → db.r5.large (-$340/ay).
4. RI/SP coverage %28 → baseline %60 plan = -$1100/ay.
5. Egress $5200 (11%):
- 60% cross-region replication ad-hoc → ZRS yeter, region azalt.
- VPC Endpoint S3 (40% NAT GW egress azalır).
6. Idle: 12 unattached EBS ($240), 3 unused EIP ($11), 2 idle ALB ($60).
7. Image registry: untagged 4.2 GB → lifecycle 7d.
8. Anomaly alert: yok → AWS Cost Anomaly Detection setup.
9. FinOps maturity: Crawl-Walk → Walk hedef (6 ay).
10. Output: 18 action item, $2400/ay tasarruf potansiyeli (= %5 toplam).
# Cost Review: <account / scope>
## Spend Snapshot
- 30d / 90d / forecast / trend
## FinOps Maturity
- Phase + hedef
## Critical / High / Medium / Low
## Right-Sizing
- compute / DB / storage
## RI / SP Coverage
- mevcut % / hedef / saving potansiyel
## Egress
- breakdown + mitigation
## Idle Resources
- liste + saving potansiyel
## Anomaly + Budget
- alert kurulumu
## Action Plan
| P | Aksiyon | Saving | Sahip | Bitiş |
| Quick (30d) | ... |
| Medium (90d) | ... |
| Strategic (1y) | ... |
## Toplam Saving Potansiyeli
- $X/ay (= %Y toplam)