az-cost-optimize Skill (awesome-copilot) for Claude Code: What It Does and How to Install It

Makes your agent find Azure savings in your resources and infrastructure code, then file GitHub issues for each fix after you say yes.

Installs: 8.7K; on skills.sh
Stars: 40.7K; on GitHub
Updated: Oct 2026; last repo commit

In seven steps the agent loads Azure best practices, discovers subscriptions, resource groups and resources, collects usage metrics, validates current costs, ranks recommendations by a priority score, and asks you for a y/n before creating issues. It then creates one issue per optimization and one EPIC issue to track them, labelled cost-optimization and azure.

It works from infrastructure-as-code files and says to stop and report if none are found. It needs an Azure MCP server and a GitHub MCP server, both signed in, and prefers the azmcp tools over the Azure CLI. It bundles no scripts.

Good for

Teams with Azure resources defined in Bicep, Terraform or ARM templates
Engineers who want savings tracked as GitHub issues
FinOps owners who want evidence-based estimates

Install az-cost-optimize

Claude Code

Copy this prompt.

Install the az-cost-optimize skill in Claude Code. Run this command:

npx skills add https://github.com/github/awesome-copilot --skill az-cost-optimize

Then check that it loaded and tell me how to use it.

2
Paste it into Claude Code. It shows the change and asks before running it.

Requires Node.js.

The awesome-copilot README is written for GitHub Copilot and does not name Claude Code, so the Claude Code and Codex routes are the standard skill-folder routes. The azure-cloud-development plugin lists 4 skills and 7 agents in its plugin.json. Six of the agents list a command-running tool and all list file editing, so review them before installing. The skill needs both MCP servers installed and signed in: the Azure server (run az login first) and the GitHub server. There is no Claude app upload route.

Built from the package data we hold. Not yet tested by us. Source.

What's in the skill

When it loads

Analyze Azure resources used in the app (IaC files and/or resources in a target rg) and optimize costs - creating GitHub issues for identified optimizations.

Sections in SKILL.md

Prerequisites
Workflow Steps
Step 1: Get Azure Best Practices
Step 2: Discover Azure Infrastructure
Step 3: Collect Usage Metrics & Validate Current Costs
Step 4: Generate Cost Optimization Recommendations
Step 5: User Confirmation
Step 6: Create Individual Optimization Issues
Step 7: Create EPIC Coordinating Issue
Error Handling
Success Criteria

Key facts

Author: Plugin author listed as Awesome Copilot Community
License: MIT (LICENSE file of the awesome-copilot repo)
Installs on skills.sh: 8,665 (Nov 2, 2026)
GitHub stars: 40,666 for the whole github/awesome-copilot repo (Nov 2, 2026)
Last repo commit: Oct 30, 2026
What it includes: 1 file: SKILL.md. No scripts
Audits on skills.sh: Gen Agent Trust Hub: pass, Socket: pass, Snyk: pass (Nov 2, 2026)
Part of: The azure-cloud-development plugin (4 skills and 7 agents listed in plugin.json)
Needs: An Azure MCP server and a GitHub MCP server, both configured and authenticated, and a target GitHub repository

What it includes

Runs azmcp-bestpractices-get before analysis
Runs code · Step 1: Get Azure best practices
Lists subscriptions, resource groups and resources, and scans for IaC files
Uses azmcp tools first and az CLI commands as a fallback. Uses only IaC files from the repo, no other files.
Runs code · Step 2: Discover infrastructure
Queries Log Analytics with KQL and checks current SKU costs
Includes predefined recent and errors queries and custom KQL for App Service, Cosmos DB and storage.
Runs code · Step 3: Usage metrics and costs
Applies compute, database, storage and infrastructure patterns, then scores priority
Priority Score = (Value Score x Monthly Savings) / (Risk Score x Implementation Days).
Step 4: Recommendations
Shows a summary and waits for y/n before creating issues
Only proceeds if you confirm.
Step 5: User confirmation
One GitHub issue per optimization, titled [COST-OPT] with the monthly savings
Labelled cost-optimization and azure. Body has the change, evidence, validation steps and risks.
Write · Step 6: Individual issues
One master issue with a Mermaid architecture diagram and a priority checklist
Labelled cost-optimization, azure and epic.
Write · Step 7: EPIC issue
Covers cost checks, Azure sign-in failure, no resources, GitHub failure, thin data
No resources found means an informational issue; a GitHub failure prints the recommendations to the console.
Error handling

From the official docs

Permissions

Medium risk

What it can do

Read your data
Create and change things
Delete things
Read-only mode

Watch out for

It reads your Azure subscription
Step 2 and 3 list subscriptions and resources, run Log Analytics queries and check costs, with Azure CLI commands as a fallback. Sign in with an account that has only reader access if you can. github.com
It creates GitHub issues with resource details
The issue template includes current configuration, usage evidence, costs and Azure CLI or deployment commands. Step 5 asks for a y/n first, and the error handling also says to create an informational issue when no resources are found. Use a private repo if the details are sensitive. github.com
The Copilot plugin route adds seven agents
plugin.json for azure-cloud-development lists 4 skills and 7 agent files. Six of the agents list a command-running tool (runCommands or execute/runInTerminal) and all seven list edit/editFiles, so read them before you install the plugin. github.com

Technical details

Turning on read-only	No read-only mode. Azure reads are separate from issue creation, which waits for your confirmation. The Azure MCP server has its own read-only option.
Talks to	Azure (through the Azure MCP server and the az CLI). GitHub (through the GitHub MCP server). azure.microsoft.com pricing pages
Scopes	Reads Azure subscriptions, resource groups, resources, usage logs and pricing Creates GitHub issues in the repository you choose Reads IaC files in the repository

Questions

What does az-cost-optimize do?

It makes your agent analyze Azure resources and IaC files for cost savings. It then creates a GitHub issue for each recommendation and one EPIC issue to coordinate them.

How do I install it?

Use the install box at the top of this page. You also need an Azure MCP server and a GitHub MCP server, because the skill calls both. In GitHub Copilot CLI it comes with the azure-cloud-development plugin.

Does it change my Azure resources?

SKILL.md describes reading Azure data and creating GitHub issues. The issues contain suggested Azure CLI and deployment commands for you or your team to run later; SKILL.md does not describe the skill running them.

Will it create issues without asking?

Step 5 shows a summary and asks "Proceed with creating GitHub issues? (y/n)" and only proceeds on a yes. The error handling section separately says to create an informational issue if no resources are found.

What if I have no IaC files?

SKILL.md says to stop and report that no IaC files were found. It treats IaC files as the only source of truth from the repo.

Does it work in the Claude app?

There is no Claude app route, because the skill relies on local Azure and GitHub MCP servers. The install box covers Claude Code and Codex.

Is it safe?

It bundles no scripts, but it reads your Azure subscription and writes issues that include resource details. Our advice: use read-only Azure access, a private repository, and review the Step 5 summary before you say yes.

Information

updated Nov 2, 2026

Repository: github/awesome-copilot
Folder: skills/az-cost-optimize

github/awesome-copilot skills.sh

Sources

Docs, issue trackers and discussions this page is built from.

Show all 10

awesome-copilot README · github.com
azure-cloud-development plugin.json · github.com
az-cost-optimize SKILL.md · github.com
azure-logic-apps-expert agent · github.com
azure-principal-architect agent · github.com
azure-saas-architect agent · github.com
azure-verified-modules-bicep agent · github.com
azure-verified-modules-terraform agent · github.com
terraform-azure-implement agent · github.com
terraform-azure-planning agent · github.com