majiayu000 Cloud FinOps Optimization System
Purpose
The finops-optimizer Skill provides comprehensive cloud cost analysis, optimization recommendations, and financial governance across AWS, Azure, and GCP. It identifies cost-saving opportunities (rightsizing, unused resources, Reserved Instances/Savings Plans), generates detailed FinOps reports, and sets up budget alerts and anomaly detection.
Only 12% of organizations have mature FinOps practices, leading to an average of 35% wasted cloud spending. Proper FinOps optimization can reduce cloud costs by 30-50% while maintaining or improving performance.
When to Use This Skill
Use finops-optimizer when you need to:
- Analyze cloud spending across multiple providers and accounts
- Identify cost optimization opportunities (rightsizing, unused resources, commitment discounts)
- Generate FinOps reports with visualizations and recommendations
- Setup budget alerts and anomaly detection
- Calculate ROI for optimization recommendations
- Track cost trends and forecast future spending
- Implement cost allocation and chargeback strategies
When NOT to Use This Skill
- For infrastructure provisioning (use
terraform-automatoror IaC tools) - For performance optimization without cost focus (use
performance-profiler) - For security compliance (use
security-scanner) - For one-time manual cost reviews (use cloud provider console)
- When you need real-time cost monitoring (use CloudWatch/Azure Monitor dashboards)
Supported Operations
1. analyze-costs - Analyze Cloud Spending
Analyzes cloud spending across providers and identifies patterns.
Input Parameters:
{
"operation": "analyze-costs",
"providers": ["aws", "azure", "gcp"],
"time_range": "last_30_days", // or "last_90_days", "current_month", "last_6_months"
"group_by": ["service", "region", "account"], // Optional
"include_forecasts": true,
"filters": {
"accounts": ["prod", "dev"],
"tags": {"Environment": "production"}
}
}
Output:
{
"success": true,
"analysis_timestamp": "2025-10-20T12:00:00Z",
"time_range": {
"start": "2025-09-20",
"end": "2025-10-20"
},
"total_cost": 45780.50,
"cost_by_provider": {
"aws": {
"total": 32500.00,
"percentage": 71.0,
"top_services": [
{
"service": "EC2",
"cost": 12500.00,
"percentage": 38.5,
"trend": "increasing"
},
{
"service": "RDS",
"cost": 8200.00,
"percentage": 25.2,
"trend": "stable"
},
{
"service": "S3",
"cost": 3800.00,
"percentage": 11.7,
"trend": "increasing"
}
]
},
"azure": {
"total": 10280.50,
"percentage": 22.5,
"top_services": [
{
"service": "Virtual Machines",
"cost": 5600.00,
"percentage": 54.5
},
{
"service": "Storage",
"cost": 2100.00,
"percentage": 20.4
}
]
},
"gcp": {
"total": 3000.00,
"percentage": 6.5,
"top_services": [
{
"service": "Compute Engine",
"cost": 1800.00,
"percentage": 60.0
}
]
}
},
"cost_by_environment": {
"production": 38500.00,
"development": 5280.50,
"staging": 2000.00
},
"cost_trends": {
"daily_average": 1526.02,
"weekly_trend": "increasing",
"month_over_month_change": "+12.5%",
"cost_velocity": "+385 USD/day"
},
"forecast": {
"next_30_days": 52000.00,
"confidence": 85,
"projected_month_end": 48500.00
},
"anomalies_detected": [
{
"date": "2025-10-15",
"service": "EC2",
"expected_cost": 400.00,
"actual_cost": 1200.00,
"deviation": "+200%",
"severity": "high"
}
]
}
Analysis Capabilities:
- Multi-Cloud: Unified analysis across AWS, Azure, GCP
- Time-Series Analysis: Trend detection and forecasting
- Anomaly Detection: Identify unusual spending patterns
- Cost Attribution: Tag-based allocation and chargeback
- Service Breakdown: Granular cost analysis by service type
2. optimize-resources - Identify Optimization Opportunities
Identifies cost-saving opportunities through resource optimization.
Input Parameters:
{
"operation": "optimize-resources",
"providers": ["aws", "azure", "gcp"],
"optimization_types": [
"rightsizing",
"unused_resources",
"reserved_instances",
"storage_optimization",
"network_optimization"
],
"minimum_savings": 100, // USD per month
"risk_tolerance": "moderate" // "conservative", "moderate", "aggressive"
}
Output:
{
"success": true,
"total_potential_savings": {
"monthly": 15680.00,
"annual": 188160.00,
"percentage_of_total": 34.3
},
"recommendations": [
{
"id": "rec-001",
"type": "rightsizing",
"provider": "aws",
"resource": "i-0abcd1234 (EC2 instance)",
"current_config": {
"instance_type": "m5.2xlarge",
"monthly_cost": 280.00,
"utilization": {
"cpu": 15.2,
"memory": 22.5
}
},
"recommended_config": {
"instance_type": "m5.large",
"monthly_cost": 70.00,
"expected_utilization": {
"cpu": 60.0,
"memory": 90.0
}
},
"potential_savings": {
"monthly": 210.00,
"annual": 2520.00,
"percentage": 75.0
},
"confidence": 95,
"risk_level": "low",
"implementation_effort": "low",
"recommendation": "Downsize from m5.2xlarge to m5.large - utilization data shows significant over-provisioning",
"migration_steps": [
"Create AMI of current instance",
"Launch m5.large with same configuration",
"Test application performance",
"Update DNS/load balancer",
"Terminate old instance"
]
},
{
"id": "rec-002",
"type": "unused_resources",
"provider": "aws",
"resource": "vol-xyz789 (EBS volume)",
"current_config": {
"size": "500 GB",
"type": "gp3",
"monthly_cost": 40.00,
"attached_to": null,
"last_attachment": "2025-07-15"
},
"potential_savings": {
"monthly": 40.00,
"annual": 480.00,
"percentage": 100.0
},
"confidence": 100,
"risk_level": "low",
"recommendation": "Delete unattached EBS volume - unused for 3 months",
"action": "Create snapshot before deletion for safety"
},
{
"id": "rec-003",
"type": "reserved_instances",
"provider": "aws",
"resource": "RDS Database Cluster",
"current_config": {
"instance_count": 3,
"instance_type": "db.r5.xlarge",
"monthly_cost_on_demand": 2100.00,
"annual_cost_on_demand": 25200.00
},
"recommended_config": {
"commitment_type": "Reserved Instance - 3 year, All Upfront",
"monthly_cost_amortized": 1260.00,
"annual_cost_amortized": 15120.00,
"upfront_payment": 45360.00
},
"potential_savings": {
"monthly": 840.00,
"annual": 10080.00,
"three_year_total": 30240.00,
"percentage": 40.0
},
"roi": {
"payback_period_months": 4.5,
"break_even_months": 54,
"net_present_value": 28500.00
},
"confidence": 90,
"risk_level": "low",
"recommendation": "Purchase 3-year Reserved Instances for stable RDS workload"
},
{
"id": "rec-004",
"type": "storage_optimization",
"provider": "azure",
"resource": "Storage Account: proddata",
"current_config": {
"tier": "Hot",
"size": "15 TB",
"monthly_cost": 3300.00,
"access_frequency": "5% last 90 days"
},
"recommended_config": {
"tier": "Cool",
"size": "15 TB",
"monthly_cost": 1650.00
},
"potential_savings": {
"monthly": 1650.00,
"annual": 19800.00,
"percentage": 50.0
},
"confidence": 85,
"risk_level": "low",
"recommendation": "Move infrequently accessed data to Cool tier",
"considerations": [
"Access costs will increase slightly",
"Overall savings remain significant",
"Implement lifecycle policies for automation"
]
}
],
"summary_by_type": {
"rightsizing": {
"count": 23,
"potential_monthly_savings": 6200.00
},
"unused_resources": {
"count": 47,
"potential_monthly_savings": 2800.00
},
"reserved_instances": {
"count": 8,
"potential_monthly_savings": 4500.00
},
"storage_optimization": {
"count": 12,
"potential_monthly_savings": 2180.00
}
},
"implementation_roadmap": {
"quick_wins": {
"count": 35,
"monthly_savings": 3200.00,
"effort": "1-2 days"
},
"medium_effort": {
"count": 30,
"monthly_savings": 7800.00,
"effort": "1-2 weeks"
},
"high_effort": {
"count": 25,
"monthly_savings": 4680.00,
"effort": "2-4 weeks"
}
}
}
Optimization Types:
- Rightsizing: Reduce over-provisioned resources
- Unused Resources: Identify and eliminate waste
- Commitment Discounts: Reserved Instances, Savings Plans, CUDs
- Storage Optimization: Tiering, lifecycle policies, compression
- Network Optimization: Data transfer, CDN usage
3. generate-report - Generate FinOps Report
Generates comprehensive FinOps report with visualizations.
Input Parameters:
{
"operation": "generate-report",
"report_type": "executive", // or "detailed", "technical", "monthly"
"time_range": "last_30_days",
"format": "markdown", // or "html", "pdf", "json"
"include_visualizations": true,
"output_file": "/tmp/finops-report.md",
"sections": [
"executive_summary",
"cost_analysis",
"optimization_opportunities",
"budget_tracking",
"recommendations",
"forecast"
]
}
Output:
{
"success": true,
"report_path": "/tmp/finops-report.md",
"report_url": "file:///tmp/finops-report.md",
"summary": {
"total_pages": 25,
"sections_included": 6,
"charts_generated": 12,
"recommendations_count": 47
},
"key_metrics": {
"total_monthly_spend": 45780.50,
"projected_annual_spend": 549366.00,
"potential_annual_savings": 188160.00,
"current_waste_percentage": 34.3,
"optimization_score": 6.5
}
}
Report Includes:
- Executive Summary: Key metrics, trends, and recommendations
- Cost Analysis: Detailed breakdown by service, region, account, tags
- Optimization Opportunities: Prioritized list with ROI calculations
- Budget Tracking: Actual vs. budgeted spend, variance analysis
- Trend Analysis: Historical spending patterns and forecasts
- Recommendations: Actionable steps with implementation roadmaps
- Visualizations: Charts for cost distribution, trends, forecasts
4. setup-alerts - Configure Budget Alerts
Sets up budget alerts and anomaly detection.
Input Parameters:
{
"operation": "setup-alerts",
"provider": "aws", // or "azure", "gcp"
"alert_configs": [
{
"name": "monthly-budget-alert",
"type": "budget",
"threshold": 50000,
"period": "monthly",
"notifications": {
"email": ["finops-team@company.com"],
"slack": "#finops-alerts"
},
"thresholds": [
{"percentage": 80, "severity": "warning"},
{"percentage": 90, "severity": "critical"},
{"percentage": 100, "severity": "emergency"}
]
},
{
"name": "anomaly-detection",
"type": "anomaly",
"sensitivity": "medium",
"services": ["EC2", "RDS", "Lambda"],
"notifications": {
"email": ["devops-team@company.com"]
}
}
]
}
Output:
{
"success": true,
"alerts_configured": [
{
"alert_id": "alert-12345",
"name": "monthly-budget-alert",
"type": "budget",
"status": "active",
"provider": "aws",
"configuration": {
"budget_amount": 50000,
"current_spend": 38500,
"percentage_used": 77.0,
"forecast_to_exceed": false
},
"notification_channels": 2
},
{
"alert_id": "alert-67890",
"name": "anomaly-detection",
"type": "anomaly",
"status": "active",
"provider": "aws",
"configuration": {
"sensitivity": "medium",
"monitored_services": ["EC2", "RDS", "Lambda"],
"detection_algorithm": "ML-based",
"lookback_period": "30 days"
}
}
],
"estimated_alert_cost": {
"monthly": 5.00,
"description": "Cost of CloudWatch alarms and SNS notifications"
}
}
Alert Types:
- Budget Alerts: Threshold-based notifications
- Anomaly Detection: ML-powered unusual spending detection
- Forecast Alerts: Predictive overspend warnings
- Service-Specific Alerts: Per-service cost monitoring
- Tag-Based Alerts: Cost center or project-level alerts
5. recommend-savings-plans - Analyze Commitment Discounts
Recommends Reserved Instances, Savings Plans, and Committed Use Discounts.
Input Parameters:
{
"operation": "recommend-savings-plans",
"providers": ["aws", "azure", "gcp"],
"commitment_types": [
"reserved_instances",
"savings_plans",
"committed_use_discounts"
],
"commitment_terms": ["1_year", "3_year"],
"payment_options": ["all_upfront", "partial_upfront", "no_upfront"],
"minimum_roi": 15 // percentage
}
Output:
{
"success": true,
"total_potential_savings": {
"annual": 125400.00,
"three_year": 376200.00
},
"recommendations": [
{
"id": "sp-001",
"provider": "aws",
"type": "Compute Savings Plan",
"commitment_type": "3_year",
"payment_option": "all_upfront",
"hourly_commitment": 5.25,
"upfront_cost": 138060.00,
"current_annual_cost": 184500.00,
"savings_plan_annual_cost": 138060.00,
"annual_savings": 46440.00,
"savings_percentage": 25.2,
"roi": {
"year_1": 33.6,
"year_3": 100.8,
"payback_period_months": 36,
"net_present_value": 128500.00
},
"coverage": {
"instances_covered": 25,
"utilization_estimate": 98.5
},
"risk_assessment": {
"level": "low",
"factors": [
"Stable workload for 18+ months",
"High utilization (98.5%)",
"Long-term business commitment"
]
},
"recommendation": "High confidence - stable compute workload justifies 3-year commitment"
},
{
"id": "ri-002",
"provider": "azure",
"type": "Reserved VM Instances",
"commitment_type": "1_year",
"payment_option": "monthly",
"instance_details": {
"size": "Standard_D4s_v3",
"quantity": 10,
"region": "East US"
},
"current_annual_cost": 52800.00,
"reserved_annual_cost": 36960.00,
"annual_savings": 15840.00,
"savings_percentage": 30.0,
"roi": {
"year_1": 30.0,
"payback_period_months": 12
},
"recommendation": "Medium confidence - consider 1-year term for flexibility"
}
],
"portfolio_optimization": {
"current_commitment_coverage": 45.0,
"recommended_coverage": 75.0,
"on_demand_remaining": 25.0,
"blended_savings_rate": 28.5,
"total_upfront_investment": 215000.00,
"break_even_date": "2026-04-15"
}
}
Commitment Types:
- AWS: Reserved Instances, Savings Plans (Compute, EC2, SageMaker)
- Azure: Reserved VM Instances, Reserved Capacity
- GCP: Committed Use Discounts (CUDs)
6. detect-anomalies - Detect Cost Anomalies
Detects unusual spending patterns using ML algorithms.
Input Parameters:
{
"operation": "detect-anomalies",
"providers": ["aws", "azure", "gcp"],
"time_range": "last_30_days",
"sensitivity": "medium", // "low", "medium", "high"
"minimum_impact": 50, // USD
"include_forecasts": true
}
Output:
{
"success": true,
"anomalies_detected": 8,
"total_anomalous_spend": 4850.00,
"anomalies": [
{
"id": "anom-001",
"date": "2025-10-15",
"provider": "aws",
"service": "EC2",
"region": "us-east-1",
"expected_cost": 400.00,
"actual_cost": 1200.00,
"deviation": 800.00,
"deviation_percentage": 200.0,
"severity": "high",
"confidence": 95,
"root_cause_analysis": {
"likely_causes": [
"New instance launched: i-0xyz123",
"Instance type: m5.8xlarge (high cost)",
"Launch time: 2025-10-15 08:23:00 UTC"
],
"resource_details": {
"instance_id": "i-0xyz123",
"instance_type": "m5.8xlarge",
"hourly_cost": 1.536,
"runtime_hours": 520
},
"recommendations": [
"Verify if m5.8xlarge is required",
"Consider rightsizing to m5.4xlarge",
"Enable auto-scaling if applicable"
]
},
"impact": {
"monthly_impact": 800.00,
"annual_impact": 9600.00
},
"status": "needs_investigation",
"assigned_to": null
},
{
"id": "anom-002",
"date": "2025-10-18",
"provider": "aws",
"service": "Data Transfer",
"region": "eu-west-1",
"expected_cost": 150.00,
"actual_cost": 850.00,
"deviation": 700.00,
"deviation_percentage": 466.7,
"severity": "critical",
"confidence": 92,
"root_cause_analysis": {
"likely_causes": [
"Unusual data transfer volume",
"Potential data exfiltration or misconfiguration",
"Cross-region replication enabled"
],
"recommendations": [
"Review VPC Flow Logs",
"Check for unauthorized applications",
"Verify S3 replication settings"
]
},
"impact": {
"monthly_impact": 700.00,
"annual_impact": 8400.00
},
"status": "critical_investigation_required"
}
],
"patterns_identified": [
{
"pattern": "Weekend EC2 spikes",
"frequency": "weekly",
"cost_impact": 1200.00,
"recommendation": "Implement scheduled scaling to shut down dev/test instances on weekends"
}
],
"forecast_anomalies": [
{
"projected_date": "2025-10-25",
"service": "RDS",
"expected_cost": 8200.00,
"forecasted_cost": 12500.00,
"confidence": 78,
"reason": "Current growth trend indicates overspend"
}
]
}
Detection Methods:
- Statistical Analysis: Standard deviation, moving averages
- Machine Learning: Pattern recognition, time-series forecasting
- Rule-Based: Threshold violations, business logic
- Comparative Analysis: Week-over-week, month-over-month comparisons
Configuration
Provider Credentials
AWS Configuration:
# AWS CLI credentials
aws configure
# Or set environment variables
export AWS_ACCESS_KEY_ID=xxx
export AWS_SECRET_ACCESS_KEY=xxx
export AWS_DEFAULT_REGION=us-east-1
Azure Configuration:
# Azure CLI login
az login
# Or set environment variables
export AZURE_SUBSCRIPTION_ID=xxx
export AZURE_TENANT_ID=xxx
GCP Configuration:
# GCP authentication
gcloud auth application-default login
# Or set service account
export GOOGLE_APPLICATION_CREDENTIALS=/path/to/credentials.json
Custom Configuration
Create .finopsrc.json in project root:
{
"providers": {
"aws": {
"enabled": true,
"accounts": ["prod-account", "dev-account"],
"regions": ["us-east-1", "eu-west-1"]
},
"azure": {
"enabled": true,
"subscriptions": ["prod-sub"]
},
"gcp": {
"enabled": false
}
},
"optimization": {
"minimum_savings": 100,
"risk_tolerance": "moderate",
"auto_implement": false
},
"reporting": {
"default_format": "markdown",
"include_visualizations": true,
"email_recipients": ["finops@company.com"]
},
"alerts": {
"budget_threshold": 50000,
"anomaly_sensitivity": "medium",
"notification_channels": {
"email": ["alerts@company.com"],
"slack_webhook": "https://hooks.slack.com/xxx"
}
},
"cost_allocation": {
"tag_keys": ["Environment", "Team", "Project", "CostCenter"]
}
}
Integration with Existing Skills
Works with container-validator
# Optimize container costs
finops-optimizer analyze-costs --filter="service=ECS,EKS"
container-validator optimize --cost-aware
Works with performance-profiler
# Balance performance and cost
performance-profiler analyze
finops-optimizer optimize-resources --maintain-performance-sla
Works with infrastructure-scanner
# Identify unused infrastructure
infrastructure-scanner scan
finops-optimizer optimize-resources --type=unused_resources
Token Economics
Without finops-optimizer Skill
Manual Approach (using agents):
1. User asks: "Analyze our cloud spending and find savings"
2. Claude analyzes AWS costs (8,000 tokens)
3. Analyzes Azure costs (6,000 tokens)
4. Generates optimization recommendations (10,000 tokens)
5. Creates ROI calculations (4,000 tokens)
6. Formats report (4,000 tokens)
Total: ~32,000 tokens per analysis
Time: 15-20 minutes
With finops-optimizer Skill
Automated Approach:
1. Skill metadata loaded: 50 tokens
2. User: "Analyze cloud costs and optimize"
3. Skill triggered, SKILL.md loaded: 600 tokens
4. Execute analysis + optimization: 0 tokens (code execution)
5. Return structured results: 350 tokens
Total: ~1,000 tokens per analysis
Time: 45-60 seconds
Execution: ~45ms
Token Savings: 31,000 tokens (96.9% reduction) Time Savings: 14-19 minutes (94% reduction)
ROI Calculation
Scenario: Medium organization ($500k/month cloud spend), weekly cost reviews
Without Skill:
- 52 analyses per year
- 1,664,000 tokens per year
- ~$5.00 per analysis at $3/1M tokens
- Annual cost: $260
- Time cost: 780-1,040 minutes/year (13-17 hours)
With Skill:
- 52 analyses per year
- 52,000 tokens per year
- ~$0.16 per analysis
- Annual cost: $8
- Time cost: 40-52 minutes/year
- Savings: $252 + 12-16 hours of FinOps time
Additional Value:
- 30-50% cloud cost reduction through systematic optimization
- Average organization saving: $150k-$250k annually
- Improved budget predictability and governance
- Faster decision-making on infrastructure investments
ROI: For $500k/month cloud spend, typical 35% waste = $175k/month wasted. If skill helps capture even 50% of waste = $87.5k/month savings ($1.05M annually).
Examples
Example 1: Initial Cost Analysis
User Prompt:
"Analyze our cloud spending across all providers for the last 30 days"
Skill Execution:
{
"operation": "analyze-costs",
"providers": ["aws", "azure", "gcp"],
"time_range": "last_30_days",
"include_forecasts": true
}
Result:
Cloud Spending Analysis (Sep 20 - Oct 20):
Total Spend: $45,780.50
- AWS: $32,500 (71%)
- Azure: $10,280 (22.5%)
- GCP: $3,000 (6.5%)
Top Cost Drivers:
1. EC2 Instances: $12,500 (27.3%)
2. RDS Databases: $8,200 (17.9%)
3. Azure VMs: $5,600 (12.2%)
Trends:
- Month-over-month: +12.5%
- Cost velocity: +$385/day
- Forecast (next 30d): $52,000
Anomalies Detected: 2
- Oct 15: EC2 spike +200% ($800 overspend)
- Oct 18: Data transfer spike +466% ($700 overspend)
Recommendation: Run optimize-resources to identify $15k+ monthly savings
Example 2: Optimization Opportunities
User Prompt:
"Find cost optimization opportunities with at least $100/month savings"
Skill Execution:
{
"operation": "optimize-resources",
"providers": ["aws", "azure", "gcp"],
"minimum_savings": 100,
"risk_tolerance": "moderate"
}
Result:
Found 90 optimization opportunities:
Total Potential Savings:
- Monthly: $15,680
- Annual: $188,160 (34.3% of total spend)
Top Recommendations:
1. Rightsizing (23 instances)
Savings: $6,200/month
Example: m5.2xlarge → m5.large (15% CPU util)
Effort: Low | Risk: Low
2. Unused Resources (47 items)
Savings: $2,800/month
Example: 35 unattached EBS volumes
Effort: Low | Risk: Low
3. Reserved Instances (8 recommendations)
Savings: $4,500/month
Example: RDS 3-year RI saves $840/month
Effort: Medium | Risk: Low
4. Storage Optimization (12 items)
Savings: $2,180/month
Example: Move 15TB to Cool tier (5% access rate)
Effort: Low | Risk: Low
Quick Wins (35 items, 1-2 days effort): $3,200/month
Next Steps:
1. Review top 10 recommendations
2. Create implementation roadmap
3. Setup alerts for future waste detection
Example 3: Reserved Instance Analysis
User Prompt:
"Should we purchase Reserved Instances? Calculate the ROI."
Skill Execution:
{
"operation": "recommend-savings-plans",
"providers": ["aws"],
"commitment_terms": ["1_year", "3_year"],
"minimum_roi": 15
}
Result:
Reserved Instance & Savings Plan Recommendations:
Total Potential Savings: $125,400/year
Top Recommendation:
Type: AWS Compute Savings Plan (3-year, all upfront)
Hourly Commitment: $5.25/hour
Upfront Cost: $138,060
Financial Analysis:
Current annual cost: $184,500
Savings Plan cost: $138,060/year
Annual savings: $46,440 (25.2%)
ROI Analysis:
Year 1 ROI: 33.6%
3-Year ROI: 100.8%
Payback period: 36 months
NPV: $128,500
Coverage: 25 instances (98.5% utilization)
Risk Assessment: LOW
- Stable workload for 18+ months
- High utilization pattern
- Long-term business commitment confirmed
Recommendation: HIGH CONFIDENCE
Purchase 3-year Savings Plan for compute workload.
Additional Options:
- 1-year term available (lower savings, more flexibility)
- Partial upfront reduces initial investment
- Consider Azure RIs for VMs ($15,840/year savings)
Example 4: Anomaly Detection
User Prompt:
"Detect any unusual spending patterns from the last 30 days"
Skill Execution:
{
"operation": "detect-anomalies",
"providers": ["aws", "azure"],
"time_range": "last_30_days",
"sensitivity": "medium"
}
Result:
Cost Anomaly Detection Report:
8 anomalies detected
Total anomalous spend: $4,850
Severity: 2 Critical, 3 High, 3 Medium
Critical Anomalies:
1. Data Transfer Spike (Oct 18)
Expected: $150 | Actual: $850 (+466%)
Service: AWS Data Transfer (eu-west-1)
Root Cause Analysis:
- Unusual cross-region data transfer
- Possible misconfigured S3 replication
- Monthly impact: $700
Action Required: INVESTIGATE IMMEDIATELY
□ Review VPC Flow Logs
□ Check S3 replication settings
□ Verify no unauthorized access
2. EC2 Instance Spike (Oct 15)
Expected: $400 | Actual: $1,200 (+200%)
Service: AWS EC2 (us-east-1)
Root Cause:
- New m5.8xlarge instance launched
- Running 520 hours ($800/month impact)
Recommendation:
- Verify if m5.8xlarge is required
- Consider rightsizing to m5.4xlarge
- Setup auto-scaling
Patterns Identified:
- Weekend EC2 spikes ($1,200/month waste)
→ Implement scheduled scaling
Forecasted Anomalies:
- Oct 25: RDS projected overspend of $4,300 (78% confidence)
Action Items:
1. Investigate critical anomalies within 24 hours
2. Implement scheduled scaling for dev/test
3. Setup alerts to prevent future anomalies
Error Handling
The skill gracefully handles common scenarios:
Missing Credentials
{
"success": true,
"warning": "Azure credentials not configured, skipping Azure analysis",
"providers_analyzed": ["aws", "gcp"],
"cost_analysis": {...}
}
Insufficient Permissions
{
"success": true,
"warning": "Limited permissions for Cost Explorer, using CloudWatch billing metrics",
"analysis_method": "fallback",
"cost_analysis": {...}
}
API Rate Limiting
{
"success": true,
"info": "AWS API rate limit reached, using cached data from 2 hours ago",
"cache_age": "2 hours",
"cost_analysis": {...}
}
Best Practices
1. Regular Analysis
- Run cost analysis weekly or bi-weekly
- Track trends to prevent cost creep
- Set cost budgets for each team/project
2. Optimization Strategy
- Start with quick wins (unused resources)
- Implement rightsizing for high-utilization resources
- Consider commitment discounts for stable workloads
- Balance savings with operational flexibility
3. Governance
- Implement tag-based cost allocation
- Setup budget alerts at 80%, 90%, 100%
- Enable anomaly detection for early warning
- Regular FinOps reviews with stakeholders
4. Automation
- Automate reporting and distribution
- Use lifecycle policies for storage optimization
- Implement auto-scaling for variable workloads
- Schedule dev/test resource shutdowns
5. Cross-Team Collaboration
- Share cost reports with engineering teams
- Include cost metrics in sprint planning
- Celebrate cost optimization wins
- Foster cost-aware culture
Troubleshooting
Issue: Cost data delayed or incomplete
Solution: Cloud providers have 24-48 hour delay for cost data. Use --include-estimates flag for real-time estimates.
Issue: Optimization recommendations too aggressive
Solution: Adjust risk tolerance:
{
"risk_tolerance": "conservative",
"minimum_utilization": 60
}
Issue: Can't connect to cloud provider APIs
Solution: Verify credentials and permissions:
# AWS
aws sts get-caller-identity
aws ce get-cost-and-usage --help
# Azure
az account show
az consumption usage list --help
# GCP
gcloud auth list
gcloud billing accounts list
Performance Characteristics
- Analysis Time: 30-60 seconds for 30-day period (single provider)
- Multi-Cloud Analysis: 90-120 seconds (AWS + Azure + GCP)
- Memory Usage: ~150MB for medium-sized accounts
- Execution Time: ~45ms for metadata/triggering
- Token Cost: 1,000 tokens average (vs 32,000 manual)
- Cache: Results cached for 1 hour by default
Future Enhancements
Planned features for future versions:
- AI-Powered Recommendations: ML models for smarter optimization
- Auto-Implementation: Automated resource optimization with approval workflows
- Cost Simulation: "What-if" analysis for infrastructure changes
- Sustainability Metrics: Carbon footprint and sustainability scoring
- Cross-Cloud Migration: Cost comparison for multi-cloud workload placement
- FinOps Dashboards: Real-time web dashboards with drill-down capabilities
Related Skills
container-validator: Optimize containerized workload costsperformance-profiler: Balance performance and cost optimizationinfrastructure-scanner: Identify unused infrastructure resourcesterraform-automator: Implement cost-optimized IaC changes
Related Agents
cloud-architect: For architectural cost optimization guidancedevops-engineer: For implementing optimization recommendationsdata-analyst: For advanced cost analytics and visualization
Summary
The finops-optimizer Skill provides comprehensive cloud cost management, turning spending chaos into financial discipline. By analyzing costs across providers, identifying optimization opportunities, and implementing governance, organizations can reduce cloud waste by 30-50%.
Key Benefits:
- 96.9% token reduction vs. manual analysis
- 94% time savings (minutes vs. hours)
- 30-50% cloud cost reduction through systematic optimization
- Improved budget predictability and forecasting
- Automated anomaly detection prevents cost surprises
- Only 12% have mature FinOps - massive competitive advantage
ROI: For medium organization ($500k/month spend), typical 35% waste = $175k/month wasted. Capturing 50% of waste = $1.05M annual savings. Skill enables systematic identification and capture of these savings while reducing analysis overhead by 94%.