Microsoft's official Azure skill for pre-deployment checks: validates azure.yaml, Bicep or Terraform, RBAC role assignments and managed identity permissions before an agent deploys.
Azure AI Services Skill
Microsoft's official skill for building and troubleshooting Azure AI Search, Speech, Document Intelligence, Content Understanding, Translator, Content Safety and Azure OpenAI apps, with per-service preflight and diagnostics guides.
Install
Other ways to install
About
- Routes Foundry and AI gateway requests to the dedicated Azure skills
- AI Search: vector, hybrid and semantic queries, indexers and skillsets
- Speech, Document Intelligence v4.0, Content Understanding, Translator and Content Safety
- Per-service preflight and diagnostics references before troubleshooting
- Azure MCP tools first, az CLI fallback, SDK guides for Python, TypeScript, .NET and Java
Azure AI is the skill in Microsoft's microsoft/azure-skills repository that covers the classic Azure AI services, as opposed to Microsoft Foundry projects and agents. It opens with a routing table: Foundry projects, agents, model deployments and fine-tuning go to the microsoft-foundry skill, API Management AI gateway policies go to azure-aigateway, and everything else in its scope stays here.
That scope is AI Search (classic indexes, vector, hybrid and semantic queries, indexers and skillsets), Speech (speech-to-text, text-to-speech, fast and batch transcription, SDK logs), Document Intelligence v4.0 (layout, OCR, custom extraction models and classifiers), Content Understanding analyzers, Translator with document translation and glossaries, standalone Content Safety, and direct Azure OpenAI SDK use. For each service the skill names the Azure MCP tools or az commands to use and points to a reference with preflight checks and diagnostic steps, so the agent reads the right guide before validating inputs or chasing an error. It also covers roles, private endpoints and data handling for Document Intelligence and Content Understanding, and keyless authentication across the SDKs.
SDK quick references are included for Python, TypeScript, .NET and Java depending on the service. The skill prefers the Azure MCP server and tells the user how to enable it in Claude Code or Copilot CLI when it is missing. MIT licensed; version 1.2.6 at the time of listing.
SKILL.md
The instructions your agent loads, as published by Microsoft.
Related Skills
See all Azure AI Services alternativesMicrosoft's official Azure skill for compliance and security audits: azqr best-practice scans plus Key Vault checks for expiring or expired keys, secrets and certificates.
Microsoft's official skill for Azure Blob Storage, File Shares, Queue and Table Storage and Data Lake: when to use each, access tiers, redundancy options, Azure MCP and az CLI commands, and SDK guides.
Microsoft's official skill for preparing apps for Azure with the Azure Developer CLI: writes a deployment plan first, then generates azure.yaml, Bicep or Terraform infrastructure and Dockerfiles before handing off to validation.