azure

Build scalable, secure, and reliable applications on Microsoft Azure cloud services. Use when deploying to Azure, configuring Azure compute/storage/database services, setting up Azure networking, impl

SKILL.md

GitHub

Azure Cloud Development

Purpose: Best practices for developing and deploying applications on Microsoft Azure, including compute, storage, security, and DevOps.


When to Use This Skill

  • Deploying applications to Microsoft Azure
  • Configuring Azure compute, storage, or database services
  • Setting up Azure networking and security
  • Implementing CI/CD with Azure DevOps
  • Managing Azure costs and monitoring

Prerequisites

  • Azure subscription
  • Azure CLI installed
  • Basic cloud computing concepts

Table of Contents

  1. Azure Fundamentals
  2. Compute Services
  3. Storage Services
  4. Database Services
  5. Networking
  6. Security
  7. DevOps and CI/CD
  8. Monitoring and Logging
  9. Cost Management
  10. AI Services
  11. Best Practices

Azure Fundamentals

Resource Hierarchy

Management Group
-- Subscription
 -- Resource Group
 -- Resources (VMs, Storage, etc.)

Naming Conventions

{resource-type}-{workload}-{environment}-{region}-{instance}

Examples:
vm-web-prod-eastus-001
st-data-dev-westus-001
sql-orders-staging-northeu-001

Resource Groups

  • Group resources by lifecycle
  • One resource group per environment per application
  • Apply tags for cost allocation and management
# Create resource group
az group create --name rg-myapp-prod --location eastus --tags Environment=Production Team=Engineering

AI Services

Azure OpenAI Service

  • Purpose: LLM hosting (GPT-4o, GPT-3.5-Turbo).
  • Best Practice: Use Managed Identity for authentication. Disable keys.
  • Networking: Deploy in Virtual Network with Private Endpoint.
# Create OpenAI Resource
az cognitiveservices account create \
 --name "ai-agentx-dev" \
 --resource-group "rg-agentx-dev-eastus-001" \
 --location "eastus2" \
 --kind "OpenAI" \
 --sku "S0"

Azure AI Search

  • Purpose: Vector database and retrieval engine for RAG.
  • Configuration: Standard Tier required for Semantic Search.
  • RBAC: Assign Search Index Data Contributor to the Agent Identity.

Azure Cosmos DB (NoSQL)

  • Purpose: Conversation history and agent state.
  • Feature: Use Vector Indexing for memory retrieval.
  • Consistency: Session consistency is usually sufficient for chat.

Best Practices

[PASS] DO

  • Use Managed Identity instead of connection strings
  • Enable Azure Defender for security
  • Use availability zones for high availability
  • Implement proper tagging strategy
  • Use Infrastructure as Code (Bicep/Terraform)
  • Enable diagnostic logging
  • Use Private Endpoints for PaaS services
  • Implement proper backup and DR strategy

[FAIL] DON'T

  • Store secrets in code or config files
  • Use overly permissive network rules
  • Ignore Azure Advisor recommendations
  • Skip resource locks on production resources
  • Use public endpoints for databases
  • Forget to set up cost alerts

References


Version: 1.1 Last Updated: February 11, 2026

Troubleshooting

Issue Solution
Deployment fails with permission error Check Azure RBAC role assignments and service principal permissions
High Azure costs Use Azure Cost Management, enable auto-shutdown for dev VMs, right-size resources
Resource not found errors Verify resource group, subscription, and region are correct