Bicep

Bicep Code Review Rules

Security (Critical)

• Input Validation: Always escape or validate user-provided strings using regex validation or built-in Bicep functions before using them in resource names, tags, and outputs. Outputs must never directly include untrusted input
• Comment Hygiene: Never use HTML comments (<!-- -->) or expose template variables (e.g., {{ }}) in outputs, as these can enable injection or phishing attacks
• Never hardcode secrets, connection strings, or keys
• Use Key Vault references for secrets
• Apply least privilege to managed identities
• Enable diagnostic settings for auditing
• Use private endpoints where available
• Enforce encryption at rest for all supported resources
• Validate Azure Policy compliance for resources
• Check regulatory standards compliance (HIPAA, PCI-DSS, etc.)

Parameters (Essential)

• Use parameters for values that vary between deployments
• Mark sensitive parameters with @secure() decorator
• Provide @description() for all parameters
• Use @allowed() for constrained values
• Set sensible @minLength(), @maxLength(), @minValue(), @maxValue()
• Provide default values where appropriate to reduce required inputs

Parameters (Advanced)

• Validate complex parameter types with parameter constraints, assertions, or custom validation logic
• Document parameter purpose and expected values
• Consider using parameter files for environment-specific configurations

Resource Naming (Essential)

• Use consistent naming convention
• Include environment, region, workload in names
• Use uniqueString() for globally unique names
• Follow Azure naming rules and restrictions

Resource Tagging (Essential)

• Tag all resources with standard tags (owner, environment, cost center)
• Use consistent tag naming conventions

Resource Tagging (Advanced)

• Include tags for governance (compliance, data classification)
• Define required tags in policy and enforce them
• Reference Azure naming documentation for governance tags and enforcement via policies

Modules

• Break down large templates into modules
• One module per logical resource group
• Use outputs to pass values between modules
• Store shared modules in a registry

Best Practices

• Use existing keyword to reference existing resources
• Use dependsOn only when implicit dependencies aren't enough
• Prefer symbolic names over resourceId() functions
• Use loops (for) instead of copy-paste for similar resources
• Include template metadata block with author, version, and docs
• Use meaningful error messages with assert() for validation

API Version Management

• Avoid deprecated or preview resource API versions unless justified
• Document reasons for using preview features

Outputs

• Output only values needed by other templates/scripts
• Mark sensitive outputs with @secure() (Bicep handles this)
• Include resource IDs for downstream references

Example Patterns

@description('Environment name')
@allowed(['dev', 'staging', 'prod'])
param environment string

@description('SQL admin password')
@secure()
param sqlAdminPassword string

metadata description = 'Azure Storage Account with Key Vault integration'
metadata author = 'Infrastructure Team'
metadata version = '1.0.0'

var baseName = 'myapp-${environment}-${uniqueString(resourceGroup().id)}'
var commonTags = {
  environment: environment
  owner: 'infrastructure-team'
  costCenter: 'IT-001'
}

resource storageAccount 'Microsoft.Storage/storageAccounts@2023-01-01' = {
  name: '${baseName}sa'
  location: resourceGroup().location
  tags: commonTags
  sku: { name: 'Standard_LRS' }
  kind: 'StorageV2'
  properties: {
    minimumTlsVersion: 'TLS1_2'
    supportsHttpsTrafficOnly: true
  }
}