azure

Overview

The azure MCP server is a Model Context Protocol (MCP) server that enables AI assistants and agents to interact directly with Microsoft Azure resources and APIs through a unified, AI-friendly interface. It allows natural-language workflows to explore, query, and manage Azure infrastructure and services without requiring custom SDK integration or manual portal usage.

This server is well suited for cloud operations, infrastructure exploration, troubleshooting, and automation scenarios where Azure context needs to be embedded directly into AI-driven workflows.

Transport

stdio

Tools

• acr
• advisor
• aks
• appConfig
• appLens
• appService
• applicationInsights
• Authorization
• azureBestPractices
• azureIsv
• azureMigrate
• azureTerraformBestPractices
• bicepSchema
• cloudArchitect
• communication
• confidentialLedger
• cosmos
• deply
• eventGrid
• eventHubs
• extension
• foundry
• functionApp
• grafana
• keyVault
• Kusto
• loadTesting
• managedLustre
• marketplace
• monitor
• mySql
• policy
• postgres
• quota
• redis
• resourceHealth
• search
• serviceBus
• signalR
• speech
• sql
• storage
• storageSync
• virtualDesktop
• wrookbooks
• oneLake
• publicApi

Key Capabilities

• Azure resource discovery — Explore subscriptions, resource groups, and deployed services programmatically.
• Direct Azure API access — Execute Azure Resource Manager (ARM) operations through structured MCP calls.
• Infrastructure insight — Retrieve configuration and metadata for compute, networking, storage, and other Azure services.
• Operational automation — Enable AI-driven workflows for inspection, validation, and management of cloud resources.
• Secure, permission-aware access — Respect Azure RBAC roles and scopes so agents only see and act on authorized resources.

How It Works

The azure MCP server runs as a local or containerized MCP service and connects to Azure using standard Azure authentication mechanisms such as Azure CLI credentials, managed identities, or service principals. Once authenticated, the server exposes Azure management operations as MCP tools that AI clients can invoke.

When an agent calls a tool, the server translates the MCP request into an Azure Resource Manager API call, executes it on behalf of the user, and returns structured results over the MCP protocol. This abstraction allows AI assistants to reason about Azure environments — including topology, configuration, and state — without embedding Azure-specific SDK logic.