Azure Resource Group Analyzer

A skill for performing comprehensive analysis of Azure resource groups and generating detailed infrastructure reports. This skill uses Azure CLI commands to gather information about all deployed resources and produces a markdown report with findings, security assessments, and recommendations.

When to Use This Skill

• •User asks to "analyze a resource group" or "audit Azure resources"
• •User wants a "resource inventory" or "infrastructure report"
• •User needs to "review Azure deployment" or "assess Azure infrastructure"
• •User asks "what's deployed in my resource group"
• •User wants to understand their Azure resource landscape
• •User needs security and cost assessment of Azure resources

Prerequisites

• •Azure CLI installed and available in PATH
• •User must be logged in to Azure (az login completed)
• •Appropriate permissions to read resources in the target resource group
• •The resource group must exist in the subscription

Step-by-Step Workflow

Step 1: Identify the Target Resource Group

Get the resource group name from the user or context. If analyzing a specific resource group mentioned in code or configuration files (like setup-azure.sh), extract the name from there.

# Verify the resource group exists
az group show --name <RESOURCE_GROUP_NAME> --output json

Step 2: List All Resources in the Resource Group

Retrieve the complete inventory of resources:

az resource list --resource-group <RESOURCE_GROUP_NAME> --output json

This provides an overview including:

• •Resource names and types
• •Locations
• •SKUs
• •Provisioning states
• •Tags
• •Creation/modification timestamps

Step 3: Gather Detailed Information for Each Resource Type

For each resource type found, run the appropriate detail command. IMPORTANT: Always use --output json and append 2>/dev/null or handle errors gracefully. When prompted for confirmations, auto-accept with --yes or -y flags where applicable.

Container Apps

az containerapp show --name <NAME> --resource-group <RG> --output json
az containerapp env show --name <ENV_NAME> --resource-group <RG> --output json

Container Registry

az acr show --name <NAME> --resource-group <RG> --output json
az acr repository list --name <NAME> --output json 2>/dev/null

Cosmos DB

az cosmosdb show --name <NAME> --resource-group <RG> --output json

PostgreSQL Flexible Server

az postgres flexible-server show --name <NAME> --resource-group <RG> --output json

Azure AI Services / Cognitive Services

az cognitiveservices account show --name <NAME> --resource-group <RG> --output json

Azure Load Testing

az load show --name <NAME> --resource-group <RG> --output json

Azure Managed Grafana

# Auto-install extension if prompted
az config set extension.dynamic_install_allow_preview=true 2>/dev/null
az config set extension.use_dynamic_install=yes_without_prompt 2>/dev/null
az grafana show --name <NAME> --resource-group <RG> --output json

Log Analytics Workspaces

az monitor log-analytics workspace list --resource-group <RG> --output json

Managed Identities

These are captured in the initial resource list - no additional command needed.

App Insights

az monitor app-insights component show --app <NAME> --resource-group <RG> --output json 2>/dev/null
# OR use REST API if extension not available
az rest --method get --uri "/subscriptions/<SUB_ID>/resourceGroups/<RG>/providers/Microsoft.Insights/components/<NAME>?api-version=2020-02-02" --output json 2>/dev/null

Step 4: Generate the Analysis Report

Create a comprehensive markdown report with these sections:

Report Structure

# Azure Resource Analysis Report

**Resource Group:** `<name>`
**Location:** <location>
**Report Generated:** <date>
**Subscription ID:** `<subscription_id>`

---

## Executive Summary
[High-level overview with key findings table showing resource health status]

## Resource Inventory
[Organized by resource category with detailed tables]

### 1. Compute & Container Resources
### 2. Container Registry
### 3. Database Resources
### 4. Monitoring & Observability
### 5. AI Services
### 6. Load Testing
### 7. Managed Identities

## Architecture Diagram (Text)
[ASCII diagram showing resource relationships]

## Security Assessment
### ✅ Strengths
### ⚠️ Areas for Improvement
### 🔐 Recommendations

## Issues & Recommendations
### 🔴 Critical Issues
### 🟡 Warnings
### 🟢 Optimization Opportunities

## Cost Considerations
[Estimated monthly costs and optimization tips]

## Resource Timeline
[Creation dates of resources]

## Summary
[Key takeaways and immediate actions required]

Step 5: Return the Report in Chat

IMPORTANT: Do NOT create a markdown file. Instead, return the complete analysis report directly in the chat response. The user can copy/save it themselves if needed.

Key Information to Extract Per Resource Type

Container Apps

• •FQDN and ingress configuration
• •Current revision and image
• •CPU/Memory allocation
• •Min/Max replicas
• •Environment variables (names only, not values)
• •Running status

Databases (Cosmos DB, PostgreSQL)

• •Provisioning state (CRITICAL - flag if Failed)
• •Server state (Running/Stopped)
• •SKU/Tier
• •Storage configuration
• •Backup policies
• •Network access settings
• •Authentication methods

Container Registry

• •Login server
• •SKU tier
• •Admin enabled status
• •Repository list
• •Retention policies

AI Services

• •Endpoint URLs
• •SKU
• •Available APIs/Capabilities
• •Authentication mode

Monitoring Resources

• •Workspace IDs
• •Retention periods
• •Pricing tiers

Auto-Accept Configurations

Before running commands, set these to avoid interactive prompts:

# Disable extension install prompts
az config set extension.use_dynamic_install=yes_without_prompt 2>/dev/null
az config set extension.dynamic_install_allow_preview=true 2>/dev/null

# Set default output format
az config set core.output=json 2>/dev/null

Error Handling

• •If a command fails, log the error and continue with other resources
• •Use 2>/dev/null to suppress error output for optional queries
• •Check exit codes and provide fallback messages
• •Never let a single resource failure stop the entire analysis

Status Indicators

Use these emoji indicators in the report:

• •✅ Healthy/Succeeded
• •⚠️ Warning/Attention needed
• •⛔ Stopped/Disabled
• •🔴 Critical/Failed

Security Flags to Check

1. •Public Network Access - Flag if enabled on sensitive resources
2. •Local Authentication - Note if disabled (good) or enabled
3. •Encryption - Platform vs Customer managed keys
4. •Private Endpoints - Presence or absence
5. •Zone Redundancy - Enabled or disabled
6. •Managed Identities - System vs User assigned

Cost Estimation Guidelines

Provide rough estimates based on:

• •Container Apps: Consumption-based (~$0-50/month for light usage)
• •ACR Basic: ~$5/month
• •PostgreSQL B1ms: ~$25/month (when running)
• •Cosmos Serverless: Pay per RU consumed
• •Grafana Standard: ~$60-90/month
• •Log Analytics: ~$2.30/GB ingested

Troubleshooting

Example Output

The report should be returned directly in the chat response as formatted markdown. Do NOT save to a file unless explicitly requested by the user.

References

• •Azure CLI Documentation
• •Azure Resource Types
• •Azure Pricing Calculator