Back to directory

Azure Agent Skills

A curated collection of Agent Skills, following the open Agent Skills standard, that give AI coding assistants structured, up-to-date guidance for Azure services drawn directly from Microsoft Learn documentation.

Azure Agent Skills

Azure Agent Skills is a curated collection of agentic skills for Azure cloud development, built on the open Agent Skills standard so they work with AI coding assistants including Claude Code, Gemini CLI, Codex CLI, GitHub Copilot, Cursor, OpenCode and others. Each skill is a folder of markdown instructions, scripts and resources that an agent loads on demand, teaching it Azure service configuration, best practices, troubleshooting steps and compliance requirements without repeating the same guidance in every conversation.

Rather than relying on retrieval to summarize raw documentation text on each request, the project extracts high-value, task-oriented capabilities from Microsoft Learn content, classifies them by purpose such as limits, troubleshooting, configuration or architecture, and reveals a skill only when it matches the user's intent. Curated bundles group related skills together for a given Azure workload.

It installs through a one-click plugin install or manually, and is licensed under CC BY 4.0.

More in Agent skills

See all 581
  • Provenance gate

    github.com

    A deterministic authorization gateway that sits between an AI agent and its tools, tracing where each parameter of a side-effect call came from, and blocking the call or asking a human when the source does not check out.

  • A markdown framework of system rules that restates Isaac Asimov's Four Laws of Robotics as a set of prioritized loss calculations for large-language-model agents to follow, published as a discovery file for agents and crawlers to read.

  • Vitruvius

    github.com

    An engineering research agent that runs a discover, read, synthesize, verify and review loop over technical questions, reading sources directly rather than inferring from titles, and publishing its own scored verifier benchmark, weaknesses included.