Azure skills
The skills supply workflows, decision...
Editor Review
The plugin joins the workflow to a real subscription, so the assistant is not limited to generic cloud advice. The skill layer uses prepare, validate, deploy, diagnostics, compliance, and quota flows to say what to do first and what to avoid. The execution layer uses Azure MCP to list resource groups, read logs, and inspect storage and messaging. Foundry answers model catalog, deployment, and evaluation questions separately. Sovereign clouds can be selected in the server command, but the local CLI must log in to that same cloud. It fits cloud developers who already edit Azure projects with a coding assistant and want a pre-deploy check plus troubleshooting against live resources.
What is Azure skills
Azure Skills is a Microsoft agent plugin on GitHub. One install packages Azure skills, the Azure MCP server, and Foundry MCP. The skills supply workflows, decision trees, and guardrails for prepare, validate, deploy, upgrade, diagnostics, compliance, quotas, cost, compute, storage, identity, and messaging. Azure MCP lets the assistant query live subscription resources, logs, and storage. Foundry MCP covers model catalog, deployment, and agent workflows. Hosts include GitHub Copilot, Claude Code, Cursor, Gemini CLI, Codex, and IntelliJ. You need a local Azure CLI login before the execution layer can read a subscription.
Basic Info
Best For
Difficulty: Advanced
Azure skills Key Features
Prepare and validate before deploy
Prepare, validate, deploy, and upgrade skills state what to check and which files must pass before a container app or other deployment starts.
Diagnostics, compliance, and quotas
Diagnostics, application monitoring, compliance, resource lookup, and quota skills investigate failures and limits instead of repeating generic cloud advice.
Cost and architecture skills
Cost, compute, resource visualization, and cloud migration skills inspect spend, compare compute choices, and map an existing architecture during a move.
Two MCP execution layers
Azure MCP reaches live resources, logs, storage, and messaging. Foundry MCP handles the model catalog, deployments, agents, and evaluations.
One plugin, several hosts
The same repository can be installed for GitHub Copilot, Claude Code, Cursor, Gemini CLI, Codex, and IntelliJ with Copilot.
Azure skills Key Advantages
- One install includes prepare, validate, and deploy skills plus Azure MCP for the live subscription
- Foundry MCP keeps model catalog, deployment, and agent workflows separate from the resource inventory
- Sovereign cloud support can target the China cloud or the US Government cloud, with the local CLI logged into the same environment
- GitHub Copilot, Claude Code, Cursor, Codex, and other listed hosts can install the same plugin
Azure skills Use Cases
Validate before deployment
A developer asks the assistant to check Azure deployment files with the validate skill before running the deployment.
Debug against the subscription
When a container app fails health checks, the assistant uses Azure MCP to read live resources and logs instead of guessing from code alone.
Check identity and storage access
When a managed identity needs to read storage, the assistant uses the identity skill to name the role and can check existing resources through the execution layer.
Frequently Asked Questions
▶How is Azure Skills different from a prompt?
The project says it is not a prompt pack. Skills provide workflows, decisions, and guardrails for prepare, validate, deploy, or diagnostics. Azure MCP and Foundry MCP list resources, read logs, and work with models and deployments. Both layers ship in one plugin.
▶What do I need before installing?
The readme asks for an Azure account or subscription, npx from Node.js, and an Azure CLI login. Deployment workflows also need the Azure Developer CLI login. Without login, the assistant cannot list your resource groups. Inside Azure, managed identity is another option.
▶How do sovereign clouds work?
The default connection is the public cloud. For the China cloud or US Government cloud, add the cloud argument to the MCP command and log the local Azure CLI, PowerShell, or Developer CLI into that same cloud. Changing only the plugin is not enough.
Latest News
View More- AI News (2026/10/20): MiniMax Launches New Generation Flash Model Min...10/20/2026
- AI News (2026/10/19): DeepSeek Launches Latest Optimized Version V4.1...10/19/2026
- AI News (2026/10/16): Kling 4.0 AI Model Fully Launched10/16/2026
- AI News (2026/10/15): SpaceXSI Launches End-to-End Voice-to-Voice Mod...10/15/2026
- AI News (2026/10/14): Google Unveils Gemini 4 Argon, Now Available to...10/14/2026
User Reviews
Real reviews and feedback from users
Related Tools
Seele AI
Seele AI is an AI-powered game development tool th...
Skywork
Skywork is Kunlun Tech's Super Agent platform buil...
Flowith
Flowith AI is a smart agent workspace built around...
Manus
Manus is an action engine that goes beyond providi...
Operator
Operator is OpenAI's first AI agent — an AI tool t...
金灵AI
Gilin AI is a professional financial research AI a...
星辰Agent
Xunfei Star Agent is an AI agent development platf...
MasterAgent
MasterAgent is the world's first L4-level multi-ag...