Install and configure Azure Skills

This article shows you how to install Azure Skills, authenticate to your Azure account, and verify the installation. After setup, you can manage Azure resources from your chat interface.

Prerequisites

Make sure you have:

Node.js Long Term Support (LTS): Download from nodejs.org or use a version manager like nvm.
Azure account: Create a free account if you don't have one.
An AI assistant: GitHub Copilot CLI, Claude Code, or another compatible platform.
Azure CLI (recommended for local development): Simplifies authentication. Install Azure CLI. Environment variables or managed identity can be used as alternatives in CI/CD pipelines.

Note

Azure Skills uses the same authentication as Azure MCP Server. If you already authenticated for Azure MCP Server, Azure Skills uses those credentials.

Authenticate to Azure

Azure Skills needs credentials to access your Azure resources. Choose the method that fits your use case.

This method is the easiest for local development and testing.

Install Azure CLI:
- Visit the Azure CLI installation guide.
- Or use a package manager: brew install azure-cli (macOS), apt-get install azure-cli (Linux).
Sign in to Azure:
```
az login
```
A browser window opens. Sign in by using your Azure account credentials.
Verify authentication:
```
az account show
```
You see your subscription details in the terminal. Azure Skills detects this authentication.

Use environment variables for scripts or pipelines where the Azure CLI isn't available. Create a service principal first.

Bash or Zsh:

export AZURE_TENANT_ID="your-tenant-id"
export AZURE_CLIENT_ID="your-client-id"
export AZURE_CLIENT_SECRET="your-client-secret"

PowerShell:

$env:AZURE_TENANT_ID = "your-tenant-id"
$env:AZURE_CLIENT_ID = "your-client-id"
$env:AZURE_CLIENT_SECRET = "your-client-secret"

Windows Command Prompt:

set AZURE_TENANT_ID=your-tenant-id
set AZURE_CLIENT_ID=your-client-id
set AZURE_CLIENT_SECRET=your-client-secret

Azure Skills detects these environment variables automatically.

Install Azure Skills

Choose the installation method for your AI assistant.

Add the marketplace (first time only):

/plugin marketplace add microsoft/azure-skills

Install the plugin:

/plugin install azure@azure-skills

Update the plugin:

/plugin update azure@azure-skills

Add the marketplace (first time only):

/plugin marketplace add microsoft/azure-skills

Install the plugin:

/plugin install azure@azure-skills

Update:

/plugin marketplace update azure-skills

Before installing Azure Skills in IntelliJ IDEA, make sure you have:

Node.js 18+ installed on your system with npx available on your PATH.
Git installed and accessible from the command line.

You can verify these prerequisites by running:

npx --version
git --version

Step 1: Install GitHub Copilot plugin

Open IntelliJ IDEA.
Go to File > Settings (on Windows/Linux) or IntelliJ IDEA > Preferences (on macOS).
Navigate to Plugins in the left sidebar.
Search for GitHub Copilot in the Marketplace tab.
Install the GitHub Copilot plugin (requires version 1.5.64-242 or higher).
Restart IntelliJ IDEA when prompted.

Step 2: Enable skills for GitHub Copilot

Open IntelliJ IDEA settings or preferences again.
Navigate to Tools > GitHub Copilot > Chat.
Select the Enable Skills checkbox.
Select Apply and OK.

Step 3: Install Azure Skills

Option 1: Azure Toolkit for IntelliJ

Install the Azure Toolkit for IntelliJ plugin from the JetBrains Marketplace.
Restart IntelliJ IDEA to complete the plugin installation.
After restarting, a notification appears offering to install Azure Skills. Select Install to add the Azure skills to your environment.
Verify the installation by opening the GitHub Copilot chat window and typing /skill:azure. This command displays all available Azure skills.

Option 2: Manual install

Open a terminal or command prompt.

Run the following command to install Azure Skills globally for GitHub Copilot:

npx skills add https://github.com/microsoft/azure-skills/tree/main/.github/plugins/azure-skills/skills -a github-copilot -g -y

Wait for the installation to complete. A confirmation message appears when the Azure skills are successfully added.

For the full list of supported hosts and installation options, see the Azure Skills repository.

Verify installation

After installation, confirm Azure Skills is ready to use.

List installed plugins:
```
/plugin list
```
The output includes azure@azure-skills and available skills.
Test with a quick command:
```
/ask List my Azure subscriptions
```
Your AI assistant queries your Azure account and displays your subscriptions.

Try Azure Skills

After you install and verify Azure Skills, try these prompts in your AI assistant's chat window:

Prepare a deployment plan for your project:

Analyze my project structure and prepare a deployment plan for deploying to Azure. What infrastructure do I need and how should I set it up?

Expected response: The assistant analyzes your codebase and generates infrastructure-as-code templates. It creates a deployment plan at .azure/plan.md and asks for your approval.

Diagnose an Azure resource health issue:

I'm getting errors from my Azure App Service. What's wrong and how do I fix it?

Expected response: The assistant inspects your app service, checks logs, and reviews recent deployments. It identifies problems and provides remediation steps.

Set up monitoring and observability:

Configure monitoring and alerting for my application. Which Application Insights features should I enable? What metrics should I track?

Expected response: The assistant proposes a monitoring strategy and generates instrumentation code. It configures Azure Alerts for critical metrics tailored to your app stack.

Supported AI assistants

You can use Azure Skills across these platforms:

GitHub Copilot CLI: Full integration with GitHub Copilot tools
Claude Code: Through the Model Context Protocol (MCP)
Other MCP-compatible AI assistants: Any tool that supports MCP

For a complete list of tools that support Azure integration, see Azure MCP Server supported tools.

Configuration options

Telemetry

By default, Azure Skills collect usage telemetry to improve the service.

To disable telemetry:

export AZURE_MCP_COLLECT_TELEMETRY=false

Troubleshooting

"Authentication failed" error

Problem: Your AI assistant can't authenticate to Azure.

Solutions:

Azure CLI: Run az login again and ensure you're authenticated.
Environment variables: Verify AZURE_TENANT_ID, AZURE_CLIENT_ID, and AZURE_CLIENT_SECRET are set correctly.
Managed identity: Confirm your Azure resource has an assigned managed identity.

"Plugin not found" error

Problem: The Azure Skills plugin isn't installed or recognized.

Solutions:

To register the marketplace, run /plugin marketplace add microsoft/azure-skills .
To install the plugin, run /plugin install azure@azure-skills.
Restart your AI assistant.
Verify Node.js is installed.

"Insufficient permissions" error

Problem: Your Azure account doesn't have the required permissions.

Solutions:

Check your Azure role-based access control (RBAC) role assignments.
Request more roles from your Azure administrator.
Use a different subscription where you have higher permissions.

"Invalid subscription" error

Problem: Azure Skills can't find or access your specified subscription.

Solutions:

To see available subscriptions, run az account list.
Verify the subscription ID or name is correct.
Check that your credentials have access to the subscription.

Feedback

Was this page helpful?

Last updated on 2026-04-24

Install and configure Azure Skills

Prerequisites

Authenticate to Azure

Install Azure Skills

Verify installation

Try Azure Skills

Supported AI assistants

Configuration options

Telemetry

Troubleshooting

"Authentication failed" error

"Plugin not found" error

"Insufficient permissions" error

"Invalid subscription" error

Related content

Feedback

Additional resources