Workflow

Step 0: Validate Prerequisites

Required Tools:

• Azure CLI installed and authenticated (az login)
• Azure CLI extensions: costmanagement, resource-graph
• Azure Quick Review (azqr) installed — See Azure Quick Review

Required Permissions:

• Cost Management Reader role
• Monitoring Reader role
• Reader role on subscription/resource group

Verification commands:

Step 1: Load Best Practices

Step 1.5: Redis-Specific Analysis (Conditional)

If the user specifically requests Redis cost optimization, use the specialized Redis reference:

Reference: Azure Redis Cost Optimization

When to use:

• User mentions “Redis”, “Azure Cache for Redis”, or “Azure Managed Redis”
• Focus is on Redis resource optimization, not general subscription analysis

💡 Note: For general subscription-wide optimization, continue with Step 2. For Redis-only analysis, follow the Redis reference document.

Step 1.6: Choose Analysis Scope (for Redis-specific analysis)

If performing Redis cost optimization, ask the user to select:

1. Specific Subscription ID
2. Subscription Name
3. Subscription Prefix (e.g., “CacheTeam”)
4. All My Subscriptions
5. Tenant-wide

Wait for user response before proceeding.

Step 1.7: Storage-Specific Analysis (Conditional)

If the user requests Storage cost optimization, load: Azure Storage Cost Optimization

Triggers: “storage account cost”, “blob storage savings”, “LRS/GRS/ZRS downgrade”, “storage lifecycle savings”, “reduce storage spending”.

For Storage-only requests, follow the Storage reference. For general optimization that includes storage, continue to Step 2.

Step 1.8: AKS-Specific Analysis (Conditional)

If the user specifically requests AKS cost optimization, use the specialized AKS reference files:

When to use AKS-specific analysis:

• User mentions “AKS”, “Kubernetes”, “cluster”, “node pool”, “pod”, or “kubectl”
• User wants to enable the AKS cost analysis add-on or namespace cost visibility
• User reports a cost spike, unusual cluster utilization, or wants budget alerts

Tool Selection:

• Prefer MCP first: Use azure__aks for AKS operations (list clusters, get node pools, inspect configuration) — it provides richer metadata and is consistent with AKS skill conventions in this repo
• Fall back to CLI: Use az aks and kubectl only when the specific operation cannot be performed via the MCP surface

Reference files (load only what is needed for the request):

• Cost Analysis Add-on — enable namespace-level cost visibility
• Anomaly Investigation — cost spikes, scaling events, budget alerts

Note: For general subscription-wide cost optimization (including AKS resource groups), continue with Step 2. For AKS-focused analysis, follow the instructions in the relevant reference file above.

Step 1.9: Choose Analysis Scope (for AKS-specific analysis)

If performing AKS cost optimization, ask the user to select their analysis scope:

Prompt the user with these options:

1. Specific Cluster Name - Analyze a single AKS cluster
2. Resource Group - Analyze all clusters in a resource group
3. Subscription ID - Analyze all clusters in a subscription
4. All My Clusters - Scan all accessible clusters across subscriptions

Wait for user response before proceeding to Step 2.

Step 2: Run Azure Quick Review

Run azqr to find orphaned resources (immediate cost savings):

Reference: Azure Quick Review

What to look for:

• Orphaned resources: unattached disks, unused NICs, idle NAT gateways
• Over-provisioned resources: excessive retention periods, oversized SKUs
• Missing cost tags

Step 3: Discover Resources

Use Azure Resource Graph for efficient cross-subscription resource discovery. See Azure Resource Graph Queries for orphaned resource detection patterns.

Step 4: Query Actual Costs

Get actual cost data from Azure Cost Management API (last 30 days). Use the Cost Query Workflow with this configuration:

Create temp/cost-query.json:

Action Required: Calculate <START_DATE> (30 days ago) and <END_DATE> (today) in ISO 8601 format.

Execute and save results to output/cost-query-result<timestamp>.json. Always include the --headers "ClientType=GitHubCopilotForAzure" header in all Cost Management API requests.

💡 Tip: Also run a cost-by-service query (grouping by ServiceName) to present the total bill breakdown alongside optimization recommendations. See examples.md.

⚠️ Warning: Sequential queries to the same scope share the per-scope rate limit (4 requests/minute). If a 429 response is received, check all x-ms-ratelimit-microsoft.costmanagement-*-retry-after headers and do not send further requests until the longest retry-after duration has elapsed.

Step 5: Validate Pricing

Fetch current pricing from official Azure pricing pages using fetch_webpage:

Key services to validate:

• Container Apps: https://azure.microsoft.com/pricing/details/container-apps/ (opens in a new tab)
• Virtual Machines: https://azure.microsoft.com/pricing/details/virtual-machines/ (opens in a new tab)
• App Service: https://azure.microsoft.com/pricing/details/app-service/ (opens in a new tab)
• Log Analytics: https://azure.microsoft.com/pricing/details/monitor/ (opens in a new tab)

Important: Check for free tier allowances — many Azure services have generous free limits.

Step 6: Collect Utilization Metrics

Query Azure Monitor for utilization data (last 14 days) to support rightsizing recommendations:

Step 7: Generate Optimization Report

Generate a report to output/costoptimizereport<YYYYMMDD_HHMMSS>.md that includes an executive summary, cost breakdown by service, free tier analysis, orphaned resources, prioritized optimization recommendations, and implementation commands. Save cost query results to output/cost-query-result<YYYYMMDD_HHMMSS>.json for audit trail, then clean up temporary files.

For the complete report template, see report-template.md.