Skill Relationships
Skill → Skill usage relationships, with cross-layer references. Same color blocks indicate the same skill form.
Skill Composition
Each skill is designed for a specific kind of caller — some only fire when you name them, others load automatically when the model judges them relevant. The split below summarises both modes at a glance.
These skills only fire when you call them by name. The agent will not load them on its own — they stay inert until invoked.
These skills are loaded automatically by the agent when it judges the context relevant. You usually don't need to call them yourself.
Skill List
Sets up AI Runway on AKS — from bare cluster to running model. Covers cluster verification, controller install, GPU assessment, provider setup, and first model deployment. Helps users quickly deploy AI inference environments on AKS clusters, supporting KAITO, vLLM, and other model serving frameworks.
Guidance for instrumenting web apps with Azure Application Insights. Provides telemetry patterns, SDK setup, and configuration references. Helps developers add Application Insights telemetry instrumentation to their apps for APM. Covers custom telemetry, distributed tracing, log collection, and diagnostic analysis configuration.
Skill for using Azure AI services: Search, Speech, OpenAI, Document Intelligence. Covers vector/hybrid/semantic search, speech-to-text, text-to-speech, transcription, and OCR document recognition. Provides MCP tools and SDK references to help developers integrate Azure AI capabilities into their applications.
Configures Azure API Management as an AI Gateway for AI models, MCP tools, and agents. Covers semantic caching, token limits, content safety, load balancing, AI model governance, MCP rate limiting, jailbreak detection, and more. Supports LLM policy configuration, cost control, and API-to-MCP conversion.
End-to-end orchestrator: from business idea, app idea, or existing app to running Azure deployment. Analyzes the app, auto-detects the right Azure services, scaffolds infrastructure code, and deploys. Includes three sub-phases: prepare (architecture planning and cost estimation), scaffold (IaC generation and self-review), deploy (IaC execution and health verification). Includes cost estimates and pre-deployment approval.
Assesses whether source code is ready to deploy to Azure — the check BEFORE infrastructure work. Evaluates build health, app completeness, dependencies and local services, stack compatibility, and deployment feasibility. Identifies deployment blockers and unsupported frameworks. Answers questions like "can my app be deployed to Azure". No infrastructure creation involved.
Assesses and migrates cross-cloud workloads to Azure with reporting and code conversion. Supports AWS Lambda→Functions, Beanstalk/Heroku/App Engine→App Service, Fargate/Kubernetes/Cloud Run/Spring Boot→Container Apps. Provides migration reports and automated code conversion to help users migrate workloads from other cloud platforms to Azure.
Azure compliance and security auditing skill. Runs comprehensive resource assessment via azqr (Azure Quick Review) and performs Key Vault expiration checks (keys, certificates, secrets). Provides best-practice evaluation, policy/compliance validation, and security posture checks. Issues classified by priority (Critical/High/Medium/Low) with remediation steps.
Azure VM/VMSS (Virtual Machine / Virtual Machine Scale Set) router skill. Covers VM creation and provisioning, SKU recommendation and pricing comparison, VMSS configuration (Flexible/Uniform orchestration), autoscaling, burstable instances, GPU and HPC workloads. Takes priority over generic Azure best practices MCP tool for VM create intents. Supports capacity reservations (CRG) and Essential Machine Management (EMM).
Azure cost management skill: query costs, forecast spending, and optimize to reduce waste. Covers cost analysis (by service/resource group/tag dimensions), spending forecasting, cost optimization recommendations (orphaned resource discovery, VM rightsizing, storage cost reduction), and AKS cost analysis. Helps users understand and control Azure spending while identifying optimization opportunities.
Executes Azure deployments for already-prepared applications. Requires existing `.azure/deployment-plan.md` (from azure-prepare) and validated status (from azure-validate). Runs `azd up`, `azd deploy`, `terraform apply`, and `az deployment` commands with built-in error recovery. Final step of the complete workflow: prepare → validate → deploy.
Debugs Azure production issues using AppLens, Azure Monitor, resource health, and safe triage. Covers App Service (high CPU, deployment failures), Container Apps, Functions, AKS, VM (RDP/SSH/black screen), networking (NSG/firewall), Event Hubs, Service Bus, and more. Provides KQL log analysis and root cause identification.
Architects and provisions enterprise Azure infrastructure from workload descriptions. For cloud architects and platform engineers planning networking, identity, security, compliance, and multi-resource topologies aligned with WAF. Generates Bicep or Terraform directly (no azd). Suitable for hub-spoke networks, multi-region DR topologies, and similar scenarios.
Plans, creates, and configures production-ready AKS clusters. Covers Day-0 checklist, SKU selection (Automatic vs Standard), networking options (private API server, Azure CNI Overlay, egress config), security, operations (autoscaling, upgrade strategy, cost analysis), and pod resource optimization (VPA recommendations, resource requests/limits).
Assesses Kubernetes workloads and cluster configuration for AKS Automatic compatibility. Identifies incompatibilities (pod resource limits, node selectors, tolerations, storage classes, etc.), generates automatic fixes, and guides migration from AKS Standard to AKS Automatic. Supports manifest validation and cluster configuration assessment.
Queries and analyzes data in Azure Data Explorer (Kusto/ADX) using KQL. Covers log analytics, telemetry processing, time series analysis, and anomaly detection. Provides KQL language reference, query patterns, time series functions, built-in anomaly detection algorithms, and best practices for ADX connection and query performance optimization.
Troubleshoots Azure Messaging SDK issues for Event Hubs and Service Bus. Covers connection failures, authentication errors, message processing issues, AMQP errors, session locks, dead-letter handling, and more. Provides targeted diagnostic steps and solutions across Python, Java, JavaScript, .NET, and other language SDKs.
Prepares azd-based Azure projects for deployment. Generates azure.yaml, infrastructure code (Bicep/Terraform), and Dockerfiles for the Azure Developer CLI (azd) workflow. Supports Function Apps (timer/Service Bus triggers), managed identity configuration, and IaC generation. First step of the complete workflow: prepare → validate → deploy.
Azure quota and usage management skill. Checks service limits, validates deployment capacity, compares cross-region quota availability, and supports quota increase requests via CLI. Emphasizes using `az quota` CLI over Portal or REST API. Provides pre-built scripts for simplified quota queries.
Assesses and improves reliability posture of Azure PaaS applications (Functions and App Service). Scans deployed resources for zone redundancy, ZRS storage, health probes, and multi-region failover. Generates feature-pivoted reliability checklists and drives staged remediation via CLI or IaC patches with user confirmation.
Lists, finds, and shows Azure resources across subscriptions and resource groups. Supports any resource type (VMs, web apps, storage, container apps, databases etc.) via Azure Resource Graph and dedicated MCP tools. Ideal for resource inventory, tag audits, orphaned resource discovery, and cross-subscription queries.
Analyzes Azure resource groups and generates detailed Mermaid architecture diagrams. Queries all resources via Azure MCP or CLI, analyzes configurations and relationships, organizes by layer (network, compute, data, security, monitoring), and generates Mermaid diagrams. Produces complete documentation with resource inventory, relationship descriptions, and recommendations.
Consultative skill for Azure Storage services (Blob, File Shares, Queue, Table, and Data Lake). Answers questions about storage account tiers, blob access tiers (hot/cool/cold/archive), and redundancy options (LRS/ZRS/GRS/GZRS). Provides MCP tool and CLI command references along with multi-language SDK quick guides.
Assesses and upgrades Azure workloads between plans, SKUs, or tiers, or modernizes Azure SDK dependencies in source code. Covers Functions Consumption→Flex Consumption, App Service→Container Apps, Azure Cache for Redis→Azure Managed Redis, and Java SDK modernization (com.microsoft.azure→com.azure). Generates assessment before automated upgrade.
Pre-deployment validation for Azure readiness. Runs after azure-prepare and before azure-deploy. Checks configuration, infrastructure (Bicep/Terraform), RBAC role assignments, managed identity permissions, and prerequisites. Executes actual validation commands, records proof, and updates deployment plan status.
Read-only analysis skill that discovers available Azure OpenAI model capacity across regions and projects. Queries quota limits, compares availability, and recommends optimal deployment locations. Outputs ranked capacity tables and hands off to preset or customize for actual deployment. No write operations.
Interactive guided deployment flow for Azure OpenAI models with full customization control. Supports model version selection, SKU choice (GlobalStandard/Standard/ProvisionedManaged/DataZoneStandard), capacity configuration, RAI content filter policy selection, and advanced options (dynamic quota, priority processing, spillover). 14-phase step-by-step deployment.
Unified entry point for Azure OpenAI model deployment with intelligent intent-based routing. Routes to three modes: preset (quick deploy to optimal region), customize (full customization), and capacity (capacity discovery). Auto-detects whether users want simple deployment, custom configuration, or capacity querying.
Provisions Microsoft Entra Agent Identity Blueprints, BlueprintPrincipals, and per-instance Agent Identities via Microsoft Graph. Configures OAuth 2.0 token exchange (fmi_path, OBO, cross-tenant). Covers complete Agent Identity lifecycle including polyglot SDK sidecar deployment and permission granting.
Guides Microsoft Entra ID app registration, OAuth 2.0 authentication, and MSAL integration. Covers app registration creation, auth configuration, API permissions setup, client credential creation (secrets/certificates/FIC), and OAuth flow implementation. Supports Portal, CLI, and IaC (Bicep) methods.
Fine-tunes models on Azure AI Foundry using SFT (supervised), DPO (preference), or RFT (reinforcement with graders). Covers dataset preparation and validation, training job submission and monitoring, grader calibration, model deployment, and evaluation. Provides Python scripts for automated data validation, training submission, monitoring, and evaluation.
Core skill for Microsoft Foundry end-to-end lifecycle management. Covers model deployment, agent creation and deployment, evaluation, fine-tuning, RBAC, quota management, and troubleshooting. Organized into 25+ sub-skills supporting both Prompt Agent and Hosted Agent types. Provides execution support via azd and Foundry MCP tools.
Intelligently deploys Azure OpenAI models to optimal regions by analyzing capacity across all available regions. Automatically checks current region first, queries all regions if needed, and deploys with GlobalStandard SKU. Auto-calculates capacity at 50% of available (min 50 TPM).
Deploys Python (Flask/Django/FastAPI) code to Azure App Service Linux. Automatically creates resource groups, App Service plans and web apps, detects framework type, configures startup commands and environment variables, and deploys via azd or az CLI. For code-only deploys, not for infrastructure provisioning.
