Azure Workload Generator

Why This Matters

The naming convention (docs/standards/naming-convention.md) requires specific inputs to generate valid resource names:

Without these inputs, you CANNOT generate compliant resource names.

Minimum Required Inputs (MUST HAVE)

Before proceeding to code generation, you MUST have explicit user confirmation for:

1. ✅ Customer Code - 2-4 character lowercase identifier
2. ✅ Workload Name - Purpose/name of this workload
3. ✅ Environment Code - d (dev), t (test), a (acceptance), p (production)
4. ✅ VM Operating System - Windows or Linux
5. ✅ Resources Needed - What resources to create (VNet, VM, disks, etc.)

Recommended Additional Inputs

These have sensible defaults but should be confirmed:

6. 📋 VNet Address Space (default: from ip-addressing-scheme.md)
7. 📋 Subnet Name and Size
8. 📋 VM SKU (default: Standard_D2s_v3)
9. 📋 Data Disk requirements (yes/no, size)
10. 📋 Tag values (CostCenter, Owner, Criticality)

STOP AND ASK

If the user's request is missing ANY of the 5 required inputs above:

STOP → Use AskUserQuestion → Gather missing inputs → THEN proceed

Example: If user says "create an exact-financials workload", you are MISSING:

• Customer code ❌
• Environment ❌
• OS type ❌
• Resources needed ❌

You MUST ask for these before generating any code.

Instructions

Phase 1: Gather Required Information (MANDATORY - NEVER SKIP)

Use the AskUserQuestion tool to collect the following information from the user:

1. Customer Code (e.g., "contoso", "fabrikam")

   • Used for directory structure and resource naming
   • Should be lowercase, alphanumeric

2. Workload Name (e.g., "webapp", "database", "api")

   • Describes the purpose of this workload
   • Used in resource naming

3. Environment Code

   • d = Development
   • t = Test
   • a = Acceptance
   • p = Production

4. VNet Address Space (e.g., "10.10.0.0/16", "172.16.0.0/16")

   • CIDR notation for the virtual network
   • Should align with IP addressing scheme (but not validated)

5. Subnet Name (e.g., "app", "data", "web")

   • Logical name for the subnet

6. Subnet Size (e.g., "/24", "/26")

   • CIDR suffix only (e.g., "/24" for 256 addresses)

7. VM Operating System

   • Windows or Linux

8. VM SKU (e.g., "Standard_D2s_v3", "Standard_B2ms")

   • Azure VM size

9. Data Disk Required

   • yes or no

10. Data Disk Size in GB (if data disk required)

    • Size in GB (e.g., 128, 256, 512)

11. Instance Number (default: "01")

    • Two-digit instance identifier per naming-convention.md
    • Default to "01" for first instance

12. Tag Values (for mandatory tags)

    • CostCenter: Department or cost allocation code
    • Owner: Team or individual responsible
    • Criticality: Low, Medium, High, Critical

Phase 2: Read Standards Documentation

Before generating code, read the following standards:

1. Read docs/standards/naming-convention.md to understand:

   • Resource naming patterns
   • Character limits
   • Tagging requirements
   • Environment codes
   • Region codes (default: weu for West Europe)

2. Reference docs/standards/ip-addressing-scheme.md for context:

   • Hub-spoke topology patterns
   • Subscription-based VNet allocation
   • Platform vs workload IP ranges

Phase 3: Generate Resource Names

Based on the naming convention standard, generate resource names following these patterns:

Resource Naming Format: <resourceType>-<workload>-<environment>-<region>-<instance>

• VNet: vnet-<workload>-<environment>-weu-<instance>

  • Example: vnet-webapp-p-weu-01

• Subnet: snet-<subnetName>-<environment>-weu-<instance>

  • Example: snet-app-p-weu-01

• VM: vm<workload><environment>weu<instance> (no hyphens, max 15 chars for Windows)

  • Example: vmwebapppweu01
  • For Linux: vm-<workload>-<environment>-weu-<instance> (up to 64 chars)

• NIC: nic-<workload>-<environment>-weu-<instance>

  • Example: nic-webapp-p-weu-01

• Data Disk (if required): disk-<workload>-data-<environment>-weu-<instance>

  • Example: disk-webapp-data-p-weu-01

• Public IP (if required): pip-<workload>-<environment>-weu-<instance>

  • Example: pip-webapp-p-weu-01

• NSG: nsg-<workload>-<environment>-weu-<instance>

  • Example: nsg-webapp-p-weu-01

Validate Name Lengths:

• Windows VM names: Max 15 characters (remove hyphens if needed)
• Linux VM names: Max 64 characters
• Other resources: Check naming-convention.md for specific limits

Phase 4: Prepare Mandatory Tags

Create a tags object with all mandatory tags:

{
  "Environment": "<d|t|a|p - full name>",
  "CostCenter": "<from user input or placeholder>",
  "Owner": "<from user input or placeholder>",
  "ManagedBy": "Terraform" or "Bicep",
  "Workload": "<workload name>",
  "Criticality": "<from user input or placeholder>"
}

Tag Values:

• Environment: "Development", "Test", "Acceptance", "Production"
• ManagedBy: "Bicep" or "Terraform" depending on file
• Use placeholders like "TO_BE_DEFINED" if user doesn't provide values

Phase 5: Create Directory Structure

Create the following directory structure under azure-workloads/:

azure-workloads/
├── Customer/
│   ├── bicep-modules/        (for reusable modules)
│   ├── terraform-modules/    (for reusable modules)
│   └── <customerCode>/       (e.g., contoso/)
│       ├── bicep/
│       │   ├── <workloadname>.bicep
│       │   └── <workloadname>.bicepparam
│       └── terraform/
│           ├── <workloadname>.tf
│           ├── variables.tf
│           └── terraform.tfvars

Use the Write tool to create directories and files.

Phase 6: Generate Bicep Code

Create two files in azure-workloads/Customer/<customerCode>/bicep/:

File 1: <workloadname>.bicep

Structure:

// Azure Spoke Workload: <workload>
// Generated by azure-workload-generator
// Customer: <customerCode>
// Environment: <environment>

targetScope = 'resourceGroup'

// ============================================================================
// PARAMETERS
// ============================================================================

@description('Location for all resources')
param location string = 'westeurope'

@description('VNet address space')
param vnetAddressSpace string

@description('Subnet address prefix')
param subnetAddressPrefix string

@description('Virtual Machine SKU')
param vmSize string

@description('Admin username for the VM')
@secure()
param adminUsername string

@description('Admin password for the VM')
@secure()
param adminPassword string

@description('Data disk size in GB (0 for no data disk)')
param dataDiskSizeGB int

@description('Resource tags')
param tags object

// ============================================================================
// VARIABLES
// ============================================================================

var vnetName = '<generated-vnet-name>'
var subnetName = '<generated-subnet-name>'
var vmName = '<generated-vm-name>'
var nicName = '<generated-nic-name>'
var nsgName = '<generated-nsg-name>'
var osDiskName = '${vmName}-osdisk'
var dataDiskName = '<generated-datadisk-name>'

// ============================================================================
// RESOURCES
// ============================================================================

// Network Security Group
resource nsg 'Microsoft.Network/networkSecurityGroups@2023-05-01' = {
  name: nsgName
  location: location
  tags: tags
  properties: {
    securityRules: [
      // Add security rules based on OS type
    ]
  }
}

// Virtual Network
resource vnet 'Microsoft.Network/virtualNetworks@2023-05-01' = {
  name: vnetName
  location: location
  tags: tags
  properties: {
    addressSpace: {
      addressPrefixes: [
        vnetAddressSpace
      ]
    }
    subnets: [
      {
        name: subnetName
        properties: {
          addressPrefix: subnetAddressPrefix
          networkSecurityGroup: {
            id: nsg.id
          }
        }
      }
    ]
  }
}

// Network Interface
resource nic 'Microsoft.Network/networkInterfaces@2023-05-01' = {
  name: nicName
  location: location
  tags: tags
  properties: {
    ipConfigurations: [
      {
        name: 'ipconfig1'
        properties: {
          subnet: {
            id: vnet.properties.subnets[0].id
          }
          privateIPAllocationMethod: 'Dynamic'
        }
      }
    ]
  }
}

// Virtual Machine
resource vm 'Microsoft.Compute/virtualMachines@2023-03-01' = {
  name: vmName
  location: location
  tags: tags
  properties: {
    hardwareProfile: {
      vmSize: vmSize
    }
    osProfile: {
      computerName: vmName
      adminUsername: adminUsername
      adminPassword: adminPassword
    }
    storageProfile: {
      imageReference: {
        // Set based on OS type (Windows or Linux)
        publisher: '<publisher>'
        offer: '<offer>'
        sku: '<sku>'
        version: 'latest'
      }
      osDisk: {
        name: osDiskName
        createOption: 'FromImage'
        managedDisk: {
          storageAccountType: 'Premium_LRS'
        }
      }
      dataDisks: dataDiskSizeGB > 0 ? [
        {
          name: dataDiskName
          diskSizeGB: dataDiskSizeGB
          lun: 0
          createOption: 'Empty'
          managedDisk: {
            storageAccountType: 'Premium_LRS'
          }
        }
      ] : []
    }
    networkProfile: {
      networkInterfaces: [
        {
          id: nic.id
        }
      ]
    }
  }
}

// ============================================================================
// OUTPUTS
// ============================================================================

output vnetId string = vnet.id
output vnetName string = vnet.name
output subnetId string = vnet.properties.subnets[0].id
output vmId string = vm.id
output vmName string = vm.name
output privateIPAddress string = nic.properties.ipConfigurations[0].properties.privateIPAddress

Key Considerations:

• Reference shared Bicep modules from bicep-modules/ directory if they exist
• Use conditional logic for data disk (only create if dataDiskSizeGB > 0)
• Set correct image reference based on OS type:
  • Windows: publisher='MicrosoftWindowsServer', offer='WindowsServer', sku='2022-datacenter-azure-edition'
  • Linux: publisher='Canonical', offer='0001-com-ubuntu-server-jammy', sku='22_04-lts-gen2'
• Include comprehensive comments
• Use latest API versions

File 2: <workloadname>.bicepparam

Structure:

using './<workloadname>.bicep'

param location = 'westeurope'
param vnetAddressSpace = '<user-provided-vnet-address>'
param subnetAddressPrefix = '<calculated-subnet-cidr>'
param vmSize = '<user-provided-vm-sku>'
param adminUsername = '<placeholder-or-keyvault-reference>'
param adminPassword = '<placeholder-or-keyvault-reference>'
param dataDiskSizeGB = <user-provided-size-or-0>
param tags = {
  Environment: '<Development|Test|Acceptance|Production>'
  CostCenter: 'TO_BE_DEFINED'
  Owner: 'TO_BE_DEFINED'
  ManagedBy: 'Bicep'
  Workload: '<workload-name>'
  Criticality: 'TO_BE_DEFINED'
}

Subnet Calculation:

• Calculate subnet CIDR from VNet address space and subnet size
• Example: VNet "10.10.0.0/16" + subnet size "/24" = "10.10.0.0/24"

Phase 7: Generate Terraform Code

Create three files in azure-workloads/Customer/<customerCode>/terraform/:

File 1: <workloadname>.tf

Structure:

# Azure Spoke Workload: <workload>
# Generated by azure-workload-generator
# Customer: <customerCode>
# Environment: <environment>

terraform {
  required_version = ">= 1.5.0"
  required_providers {
    azurerm = {
      source  = "hashicorp/azurerm"
      version = "~> 3.0"
    }
  }
}

provider "azurerm" {
  features {}
}

# ============================================================================
# LOCAL VARIABLES
# ============================================================================

locals {
  vnet_name      = "<generated-vnet-name>"
  subnet_name    = "<generated-subnet-name>"
  vm_name        = "<generated-vm-name>"
  nic_name       = "<generated-nic-name>"
  nsg_name       = "<generated-nsg-name>"
  os_disk_name   = "${local.vm_name}-osdisk"
  data_disk_name = "<generated-datadisk-name>"

  common_tags = {
    Environment  = var.environment
    CostCenter   = var.cost_center
    Owner        = var.owner
    ManagedBy    = "Terraform"
    Workload     = var.workload_name
    Criticality  = var.criticality
  }
}

# ============================================================================
# RESOURCE GROUP
# ============================================================================

# Assumes resource group already exists
data "azurerm_resource_group" "main" {
  name = var.resource_group_name
}

# ============================================================================
# NETWORK SECURITY GROUP
# ============================================================================

resource "azurerm_network_security_group" "main" {
  name                = local.nsg_name
  location            = data.azurerm_resource_group.main.location
  resource_group_name = data.azurerm_resource_group.main.name
  tags                = local.common_tags

  # Add security rules based on OS type
}

# ============================================================================
# VIRTUAL NETWORK
# ============================================================================

resource "azurerm_virtual_network" "main" {
  name                = local.vnet_name
  location            = data.azurerm_resource_group.main.location
  resource_group_name = data.azurerm_resource_group.main.name
  address_space       = [var.vnet_address_space]
  tags                = local.common_tags
}

resource "azurerm_subnet" "main" {
  name                 = local.subnet_name
  resource_group_name  = data.azurerm_resource_group.main.name
  virtual_network_name = azurerm_virtual_network.main.name
  address_prefixes     = [var.subnet_address_prefix]
}

resource "azurerm_subnet_network_security_group_association" "main" {
  subnet_id                 = azurerm_subnet.main.id
  network_security_group_id = azurerm_network_security_group.main.id
}

# ============================================================================
# NETWORK INTERFACE
# ============================================================================

resource "azurerm_network_interface" "main" {
  name                = local.nic_name
  location            = data.azurerm_resource_group.main.location
  resource_group_name = data.azurerm_resource_group.main.name
  tags                = local.common_tags

  ip_configuration {
    name                          = "ipconfig1"
    subnet_id                     = azurerm_subnet.main.id
    private_ip_address_allocation = "Dynamic"
  }
}

# ============================================================================
# VIRTUAL MACHINE
# ============================================================================

resource "azurerm_<windows|linux>_virtual_machine" "main" {
  name                = local.vm_name
  location            = data.azurerm_resource_group.main.location
  resource_group_name = data.azurerm_resource_group.main.name
  size                = var.vm_size
  admin_username      = var.admin_username
  admin_password      = var.admin_password
  tags                = local.common_tags

  network_interface_ids = [
    azurerm_network_interface.main.id
  ]

  os_disk {
    name                 = local.os_disk_name
    caching              = "ReadWrite"
    storage_account_type = "Premium_LRS"
  }

  # Set source_image_reference based on OS type
  source_image_reference {
    publisher = "<publisher>"
    offer     = "<offer>"
    sku       = "<sku>"
    version   = "latest"
  }
}

# ============================================================================
# DATA DISK (CONDITIONAL)
# ============================================================================

resource "azurerm_managed_disk" "data" {
  count                = var.data_disk_size_gb > 0 ? 1 : 0
  name                 = local.data_disk_name
  location             = data.azurerm_resource_group.main.location
  resource_group_name  = data.azurerm_resource_group.main.name
  storage_account_type = "Premium_LRS"
  create_option        = "Empty"
  disk_size_gb         = var.data_disk_size_gb
  tags                 = local.common_tags
}

resource "azurerm_virtual_machine_data_disk_attachment" "data" {
  count              = var.data_disk_size_gb > 0 ? 1 : 0
  managed_disk_id    = azurerm_managed_disk.data[0].id
  virtual_machine_id = azurerm_<windows|linux>_virtual_machine.main.id
  lun                = 0
  caching            = "ReadWrite"
}

# ============================================================================
# OUTPUTS
# ============================================================================

output "vnet_id" {
  description = "The ID of the Virtual Network"
  value       = azurerm_virtual_network.main.id
}

output "vnet_name" {
  description = "The name of the Virtual Network"
  value       = azurerm_virtual_network.main.name
}

output "subnet_id" {
  description = "The ID of the Subnet"
  value       = azurerm_subnet.main.id
}

output "vm_id" {
  description = "The ID of the Virtual Machine"
  value       = azurerm_<windows|linux>_virtual_machine.main.id
}

output "vm_name" {
  description = "The name of the Virtual Machine"
  value       = azurerm_<windows|linux>_virtual_machine.main.name
}

output "private_ip_address" {
  description = "The private IP address of the VM"
  value       = azurerm_network_interface.main.private_ip_address
}

Key Considerations:

• Use azurerm_windows_virtual_machine for Windows, azurerm_linux_virtual_machine for Linux
• Reference shared Terraform modules from terraform-modules/ directory if they exist
• Use count for conditional data disk creation
• Set correct image reference based on OS type:
  • Windows: publisher='MicrosoftWindowsServer', offer='WindowsServer', sku='2022-datacenter-azure-edition'
  • Linux: publisher='Canonical', offer='0001-com-ubuntu-server-jammy', sku='22_04-lts-gen2'
• For Windows VMs, add disable_password_authentication = false is not applicable
• For Linux VMs, consider SSH key authentication instead of password

File 2: variables.tf

Structure:

# Variable definitions for <workload> workload

variable "resource_group_name" {
  description = "Name of the resource group"
  type        = string
}

variable "vnet_address_space" {
  description = "Address space for the virtual network"
  type        = string
}

variable "subnet_address_prefix" {
  description = "Address prefix for the subnet"
  type        = string
}

variable "vm_size" {
  description = "Size of the virtual machine"
  type        = string
}

variable "admin_username" {
  description = "Admin username for the VM"
  type        = string
  sensitive   = true
}

variable "admin_password" {
  description = "Admin password for the VM"
  type        = string
  sensitive   = true
}

variable "data_disk_size_gb" {
  description = "Size of the data disk in GB (0 for no data disk)"
  type        = number
  default     = 0
}

variable "environment" {
  description = "Environment name (Development, Test, Acceptance, Production)"
  type        = string
}

variable "cost_center" {
  description = "Cost center tag value"
  type        = string
  default     = "TO_BE_DEFINED"
}

variable "owner" {
  description = "Owner tag value"
  type        = string
  default     = "TO_BE_DEFINED"
}

variable "workload_name" {
  description = "Workload name for tagging"
  type        = string
}

variable "criticality" {
  description = "Criticality level (Low, Medium, High, Critical)"
  type        = string
  default     = "TO_BE_DEFINED"
}

File 3: terraform.tfvars

Structure:

# Terraform variables for <workload> workload
# Customer: <customerCode>
# Environment: <environment>

resource_group_name    = "rg-<workload>-<environment>-weu-01"
vnet_address_space     = "<user-provided-vnet-address>"
subnet_address_prefix  = "<calculated-subnet-cidr>"
vm_size                = "<user-provided-vm-sku>"
admin_username         = "azureadmin"  # Change as needed
admin_password         = "CHANGE_ME_SECURE_PASSWORD"  # Use Key Vault or environment variable
data_disk_size_gb      = <user-provided-size-or-0>
environment            = "<Development|Test|Acceptance|Production>"
workload_name          = "<workload-name>"
cost_center            = "TO_BE_DEFINED"
owner                  = "TO_BE_DEFINED"
criticality            = "TO_BE_DEFINED"

Phase 8: Validation and Summary

After generating all files:

1. Validate that all files are created in the correct directory structure
2. Verify resource names follow naming-convention.md patterns
3. Check that all mandatory tags are included
4. Confirm that both Bicep and Terraform code are consistent
5. Review subnet CIDR calculation is correct

Provide a summary to the user:

✓ Generated Azure Workload: <workload>
✓ Customer: <customerCode>
✓ Environment: <environment>

Files Created:
- azure-workloads/Customer/<customerCode>/bicep/<workloadname>.bicep
- azure-workloads/Customer/<customerCode>/bicep/<workloadname>.bicepparam
- azure-workloads/Customer/<customerCode>/terraform/<workloadname>.tf
- azure-workloads/Customer/<customerCode>/terraform/variables.tf
- azure-workloads/Customer/<customerCode>/terraform/terraform.tfvars

Resources:
- VNet: <vnet-name> (<vnet-address-space>)
- Subnet: <subnet-name> (<subnet-cidr>)
- VM: <vm-name> (<vm-sku>, <os-type>)
- Data Disk: <Yes/No> (<size-if-yes>)

Next Steps:
1. Review and update placeholder tag values (CostCenter, Owner, Criticality)
2. Update admin credentials in parameter files
3. Create resource group: rg-<workload>-<environment>-weu-01
4. Deploy using:
   - Bicep: az deployment group create --resource-group <rg-name> --parameters <workloadname>.bicepparam
   - Terraform: terraform init && terraform plan && terraform apply

Examples

Example 1: Web Application Workload

User Input:

• Customer Code: contoso
• Workload Name: webapp
• Environment: p (Production)
• VNet Address Space: 10.10.0.0/16
• Subnet Name: app
• Subnet Size: /24
• VM OS: Linux
• VM SKU: Standard_D2s_v3
• Data Disk: yes
• Data Disk Size: 128 GB
• Instance: 01
• Tags: CostCenter=CC-12345, Owner=WebTeam, Criticality=High

Generated Resources:

• VNet: vnet-webapp-p-weu-01
• Subnet: snet-app-p-weu-01
• VM: vm-webapp-p-weu-01 (Linux, up to 64 chars allowed)
• NIC: nic-webapp-p-weu-01
• NSG: nsg-webapp-p-weu-01
• Data Disk: disk-webapp-data-p-weu-01

Files Created:

azure-workloads/Customer/contoso/
├── bicep/
│   ├── webapp.bicep
│   └── webapp.bicepparam
└── terraform/
    ├── webapp.tf
    ├── variables.tf
    └── terraform.tfvars

Example 2: Database Workload (Windows)

User Input:

• Customer Code: fabrikam
• Workload Name: sqldb
• Environment: t (Test)
• VNet Address Space: 172.16.0.0/16
• Subnet Name: data
• Subnet Size: /26
• VM OS: Windows
• VM SKU: Standard_E4s_v3
• Data Disk: yes
• Data Disk Size: 512 GB
• Instance: 01
• Tags: CostCenter=CC-67890, Owner=DataTeam, Criticality=Critical

Generated Resources:

• VNet: vnet-sqldb-t-weu-01
• Subnet: snet-data-t-weu-01
• VM: vmsqldbtweu01 (Windows, max 15 chars, no hyphens)
• NIC: nic-sqldb-t-weu-01
• NSG: nsg-sqldb-t-weu-01
• Data Disk: disk-sqldb-data-t-weu-01

Files Created:

azure-workloads/Customer/fabrikam/
├── bicep/
│   ├── sqldb.bicep
│   └── sqldb.bicepparam
└── terraform/
    ├── sqldb.tf
    ├── variables.tf
    └── terraform.tfvars

Requirements

Documentation References

• CRITICAL: Must read docs/standards/naming-convention.md before generating names
• REFERENCE: docs/standards/ip-addressing-scheme.md for network context

Azure Permissions

Users deploying the generated code need:

• Contributor role on target resource group
• Network Contributor for VNet operations

Tools Required for Deployment

• Bicep: Azure CLI 2.20.0+ with Bicep
• Terraform: Terraform 1.5.0+, Azure CLI for authentication

Best Practices

Security

• Never hardcode credentials in generated files
• Use Azure Key Vault references for secrets in production
• Implement network security group rules appropriate for workload type
• Consider using managed identities instead of passwords where possible

Naming

• Strictly follow naming-convention.md patterns
• Validate Windows VM names are ≤15 characters (remove hyphens)
• Linux VM names can be up to 64 characters (keep hyphens)
• Use instance number 01 as default, increment for additional instances

Network Design

• Ensure VNet address space doesn't overlap with existing networks
• Subnet should be appropriately sized for expected resources
• Consider future growth when selecting address spaces
• Follow ip-addressing-scheme.md for subscription-based allocation

Tagging

• All resources must have mandatory tags
• Use placeholder values if user doesn't provide tag information
• ManagedBy tag should reflect IaC tool (Bicep or Terraform)
• Environment tag should use full name, not code

Code Quality

• Include comprehensive comments in generated code
• Use latest stable API versions
• Implement conditional logic for optional resources (data disk)
• Generate consistent code across Bicep and Terraform
• Reference shared modules when available

Advanced Usage

Module References

If reusable Bicep or Terraform modules exist in the bicep-modules/ or terraform-modules/ directories, the skill should reference them instead of inline resources:

Bicep Module Reference Example:

module vnet '../../../bicep-modules/vnet/main.bicep' = {
  name: 'vnet-deployment'
  params: {
    vnetName: vnetName
    addressSpace: vnetAddressSpace
    location: location
    tags: tags
  }
}

Terraform Module Reference Example:

module "vnet" {
  source = "../../../terraform-modules/vnet"

  vnet_name     = local.vnet_name
  address_space = var.vnet_address_space
  location      = data.azurerm_resource_group.main.location
  tags          = local.common_tags
}

Multi-Instance Deployments

For deploying multiple instances of the same workload:

1. Use instance number 01, 02, 03, etc.
2. Ensure each instance has unique IP ranges
3. Consider creating separate parameter files per instance

Integration with CI/CD

The generated code can be integrated into CI/CD pipelines:

• Bicep: Use Azure DevOps or GitHub Actions with az deployment group create
• Terraform: Use Terraform Cloud, Azure DevOps, or GitHub Actions with terraform apply

Troubleshooting

Common Issues

Issue: VM name exceeds 15 characters for Windows Solution: Remove hyphens and shorten workload name. Example: vmsqldbtweu01 instead of vm-sqldb-t-weu-01

Issue: Subnet CIDR calculation incorrect Solution: Ensure subnet size suffix is applied to VNet base address correctly

Issue: IP address space conflicts Solution: Verify VNet address doesn't overlap with existing networks or hub VNets

Issue: Mandatory tags missing Solution: Ensure all six mandatory tags are included in tags object

Issue: Module references not found Solution: Check if shared modules exist in bicep-modules/ or terraform-modules/ directories

Summary

The azure-workload-generator skill automates the creation of production-ready Azure spoke workload infrastructure following enterprise standards. It generates both Bicep and Terraform code with proper:

• Resource naming per naming-convention.md
• Mandatory tagging
• Network topology aligned with ip-addressing-scheme.md
• Security best practices
• Conditional resource creation
• Comprehensive documentation

Use this skill whenever you need to create new Azure workload infrastructure for customers.