cost-optimization
installsforks
How to install cost-optimization?
thomast1906/github-copilot-agent-skills/cost-optimization
$npx -y skills add thomast1906/github-copilot-agent-skills --skill cost-optimizationAnalyze Azure architectures for cost optimization opportunities, provide savings recommendations, and calculate ROI for improvements.
SKILL.md
| 1 | # Cost Optimization Skill |
| 2 | |
| 3 | Analyze Azure costs and identify optimization opportunities across compute, storage, networking, and data services. Provide actionable recommendations with savings estimates. |
| 4 | |
| 5 | ## When to Use |
| 6 | |
| 7 | - Review architecture costs and identify waste |
| 8 | - Optimize existing Azure deployments |
| 9 | - Right-size over-provisioned resources |
| 10 | - Implement reserved instances and savings plans |
| 11 | - Set up cost monitoring and alerts |
| 12 | - Reduce monthly Azure bills |
| 13 | |
| 14 | ## Cost Optimization Categories |
| 15 | |
| 16 | ### 1. Right-Sizing |
| 17 | Adjust resource SKUs to match actual usage patterns. |
| 18 | |
| 19 | **Target Resources:** |
| 20 | - Virtual Machines |
| 21 | - App Service Plans |
| 22 | - SQL Databases |
| 23 | - Cosmos DB throughput |
| 24 | - Azure Cache for Redis |
| 25 | |
| 26 | **Analysis Method:** |
| 27 | 1. Review 30-day metrics (CPU, memory, DTU utilization) |
| 28 | 2. Identify resources with < 40% average utilization |
| 29 | 3. Recommend smaller SKU or scaling adjustments |
| 30 | |
| 31 | **Typical Savings:** 30-50% |
| 32 | |
| 33 | ### 2. Reserved Instances & Savings Plans |
| 34 | Commit to 1-year or 3-year terms for predictable workloads. |
| 35 | |
| 36 | **Eligible Services:** |
| 37 | - Virtual Machines |
| 38 | - App Service Plans |
| 39 | - Azure SQL Database |
| 40 | - Cosmos DB |
| 41 | - Azure Cache for Redis |
| 42 | |
| 43 | **Savings (typical ranges — always verify with the `azure-pricing` skill using `price-type: Reservation`):** |
| 44 | - 1-year: 20-40% |
| 45 | - 3-year: 40-72% |
| 46 | |
| 47 | **When to Use:** Workloads with consistent, predictable usage |
| 48 | |
| 49 | **To confirm exact RI rates:** use `tool_search_tool_regex` with pattern `pricing`, then call the tool with `price-type: Reservation` and `include-savings-plan: true` for the specific SKU and region. Compare the returned `retailPrice` against the `Consumption` rate to calculate the actual saving percentage. |
| 50 | |
| 51 | ### 3. Auto-Scaling |
| 52 | Scale resources based on demand instead of static provisioning. |
| 53 | |
| 54 | **Applicable Services:** |
| 55 | - App Service |
| 56 | - Virtual Machine Scale Sets |
| 57 | - Container Apps |
| 58 | - AKS node pools |
| 59 | - Cosmos DB autoscale |
| 60 | |
| 61 | **Typical Savings:** 20-40% (eliminates idle capacity during off-peak) |
| 62 | |
| 63 | ### 4. Storage Tiering |
| 64 | Move infrequently accessed data to cheaper storage tiers. |
| 65 | |
| 66 | **Blob Storage Tiers:** |
| 67 | - **Hot**: Frequent access (< 30 days old) |
| 68 | - **Cool**: Infrequent access (30-90 days), 50% cheaper |
| 69 | - **Archive**: Rare access (> 90 days), 90% cheaper |
| 70 | |
| 71 | **Implementation:** Lifecycle management policies |
| 72 | |
| 73 | **Typical Savings:** 50-90% on archived data |
| 74 | |
| 75 | ### 5. Eliminate Waste |
| 76 | Identify and remove unused resources. |
| 77 | |
| 78 | **Common Waste:** |
| 79 | - Unattached disks |
| 80 | - Stopped (but not deallocated) VMs |
| 81 | - Orphaned public IPs |
| 82 | - Unused App Service Plans |
| 83 | - Old snapshots and backups |
| 84 | - Idle Load Balancers |
| 85 | |
| 86 | **Typical Savings:** £200-2,000/month per environment |
| 87 | |
| 88 | ## Cost Analysis Process |
| 89 | |
| 90 | ### Step 0: Retrieve Live Pricing with Azure MCP Pricing Tool |
| 91 | |
| 92 | **Before estimating any costs, invoke the `azure-pricing` skill to fetch real retail prices.** |
| 93 | |
| 94 | First, use `tool_search_tool_regex` with pattern `pricing` to discover the exact tool name — do **not** hardcode it. Then call the tool with: |
| 95 | |
| 96 | ``` |
| 97 | sku: <ARM SKU e.g. Standard_D4ds_v5> |
| 98 | service: <e.g. Virtual Machines, Azure Kubernetes Service> |
| 99 | region: <ARM region slug e.g. uksouth, eastus> |
| 100 | currency: GBP |
| 101 | price-type: Consumption # on-demand baseline |
| 102 | include-savings-plan: true # returns 1yr/3yr rates in nested array |
| 103 | ``` |
| 104 | |
| 105 | Then follow up with `price-type: Reservation` to retrieve exact reserved instance rates. |
| 106 | |
| 107 | **Monthly cost formula:** `hourly_price × 730` |
| 108 | |
| 109 | **Important:** The tool requires a specific SKU or service name — do not call it with only a broad category (e.g. "Virtual Machines"). Confirm the SKU before calling. |
| 110 | |
| 111 | Build a three-column cost table per resource: |
| 112 | |
| 113 | | Resource | Pay-as-you-go | 1-yr Reserved | 3-yr Reserved | |
| 114 | |----------|--------------|---------------|---------------| |
| 115 | | (data from pricing tool) | | | | |
| 116 | |
| 117 | ### Step 1: Gather Current Costs |
| 118 | |
| 119 | Extract cost data from Azure Cost Management: |
| 120 | - Last 30-60 days of spending by resource |
| 121 | - Group by resource type and resource group |
| 122 | - Identify top 10 cost contributors |
| 123 | |
| 124 | ### Step 2: Analyze Resource Utilization |
| 125 | |
| 126 | For each major resource: |
| 127 | - **Compute**: Average CPU, memory utilization |
| 128 | - **Database**: DTU/vCore usage, storage growth |
| 129 | - **Storage**: Access patterns, growth rate |
| 130 | - **Networking**: Bandwidth usage, idle resources |
| 131 | |
| 132 | ### Step 3: Identify Opportunities |
| 133 | |
| 134 | Categorize findings: |
| 135 | - **Quick Wins**: < 1 hour, immediate savings (delete unused resources) |
| 136 | - **Right-Sizing**: < 1 day, 30-50% savings |
| 137 | - **Reserved Instances**: < 1 hour setup, 1-3 year commitment |
| 138 | - **Architecture Changes**: > 1 week, significant redesign |
| 139 | |
| 140 | ### Step 4: Calculate ROI |
| 141 | |
| 142 | For each recommendation: |
| 143 | - Current monthly cost |
| 144 | - Optimized monthly cost |
| 145 | - Monthly savings |
| 146 | - Implementation effort (hours) |
| 147 | - Break-even time |
| 148 | |
| 149 | ## Output Format |
| 150 | |
| 151 | > **Note:** All cost figures in the template below are illustrative placeholders. Before |
People who liked this also viewed
6 picksType
- azure-costAzure cost management: query costs, forecast spending, optimize to reduce waste. WHEN: \"Azure costs\", \"Azure bill\", \"cost breakdown\", \"how much am I spending\", \"forecast spending\", \"optimizSkillsJul 2026
- pricingWhen the user wants help with pricing decisions, packaging, or monetization strategy. Also use when the user mentions 'pricing,' 'pricing tiers,' 'freemium,' 'free trial,' 'packaging,' 'price increaseSkillsJul 2026
- stripe-integrationMulti-harness agentic plugin marketplace for Claude Code, Codex CLI, Cursor, OpenCode, GitHub Copilot, and Gemini CLISkillsJul 2026
- startup-financial-modelingMulti-harness agentic plugin marketplace for Claude Code, Codex CLI, Cursor, OpenCode, GitHub Copilot, and Gemini CLISkillsJul 2026
- tushare-finance专为 Claude Code 设计的 Tushare Pro 金融数据技能包，支持 220+ 个接口，用自然语言获取 A 股行情、财务数据、宏观经济等金融数据。SkillsJul 2026
- azure-pricingFetches real-time Azure retail pricing using the Azure Retail Prices API (prices.azure.com) and estimates Copilot Studio agent credit consumption. Use when the user asks about the cost of any Azure seSkillsJul 2026
