Agent-Skills.md

Agent Skills: Microsoft Skills

Use microsoft/skills for Azure SDK grounding across .NET and Java agent coding workflows.

UncategorizedID: autohandai/community-skills/microsoft-skills

Repository

autohandai/community-skills
autohandaiLicense: Apache-2.0
103

Install this agent skill to your local

pnpm dlx add-skill https://github.com/autohandai/community-skills/tree/HEAD/microsoft-skills

Skill Files

Browse the full folder contents for microsoft-skills.

Download Skill

Loading file tree…

microsoft-skills/SKILL.md

Skill Metadata

Name
microsoft-skills
Description
"Use microsoft/skills for Azure SDK grounding across .NET and Java agent coding workflows."

Microsoft Skills

Overview

Use this skill when work needs Microsoft SDK grounding for coding agents, especially Azure SDK tasks across .NET and Java. The upstream microsoft/skills repository includes skills, MCP servers, custom agents, and agent instructions for SDK-focused development.

Compatibility: Azure SDK, .NET, Java, MCP servers, custom agents, and agent instruction files.

Workflow

  1. Identify the SDK family, target language, package versions, authentication model, and Azure service.
  2. Read the relevant upstream skill or agent instructions before applying service-specific patterns.
  3. Prefer official SDK APIs and examples over hand-rolled REST calls unless the SDK lacks coverage.
  4. Validate with language-native tooling: dotnet build/test for .NET and Maven or Gradle tests for Java.
  5. Document service prerequisites, identity permissions, and environment variables separately from source changes.

Guardrails

  • Do not assume Azure credentials, subscriptions, or cloud resources are available locally.
  • Keep secret values out of code, generated docs, and command output.
  • Match the repository's existing SDK major versions and dependency management approach.
  • For migrations, preserve client behavior and retry/error handling semantics.

Expected Output

Return SDK-grounded implementation guidance with package choices, authentication assumptions, validation commands, and source references.