What does this skill do, and when should you use it?
Azure Skill Creator is an authoring guide for AI coding-agent skills, focused on Azure SDKs and Microsoft Foundry. It defines metadata, structure, progressive disclosure, token budgets, authentication, client lifecycle guidance, and reference-file organization. It also specifies required SDK context and validation checks for examples, credentials, frontmatter, and efficiency. Adopt it when you need to create or maintain SDK/API skills; it is not itself a service-specific implementation guide.
It guides authors through gathering SDK/API context, writing a YAML-frontmatter SKILL.md, selecting a core workflow, and moving advanced material into references/. It defines sections for installation, environment variables, authentication, client lifecycle, examples, feature tables, best practices, and links. It also supplies anti-pattern checks, token budgets, credential rules, frontmatter validation, and optional Vally and test-harness procedures.
- An Azure SDK engineer needs to create a coding-agent skill from a package and its official documentation.
- A maintainer needs to update an existing skill for new SDK patterns, authentication guidance, or a renamed package.
- A skill author needs to keep SKILL.md concise while moving batch operations, error handling, or advanced authentication into references/.
- A team needs a pre-release checklist for examples, credentials, metadata, token limits, and runnable workflows.
How do you install this skill?
- Do not treat the Vally, CI, or example commands as executed test evidence; separately verify dependencies, packaging paths, and authentication examples in the target environment.
- The initializer creates placeholder directories and files; review generated content and confirm the scope of any cleanup before removing unused resources.
- For mainland-China users, verify reachability of Microsoft documentation, MCP services, Azure endpoints, and the npx installation chain in advance.
- Shell / CLI
- Network access
- Local filesystem
- MCP Server
microsoft-docs MCP
Install the repository collection with npx skills add microsoft/skills, then select the required skill in the wizard. Alternatively, clone the repository and copy .github/skills/skill-creator/ into the target agent directory, such as .github/skills/ for GitHub Copilot. The source does not document a dedicated single-skill installation command.
tmp="$(mktemp -d)"
git clone --depth 1 https://github.com/microsoft/skills.git "$tmp"
mkdir -p ~/.claude/skills
cp -R "$tmp/.github/skills/skill-creator" ~/.claude/skills/
rm -rf "$tmp"Generated from the source repository and skill path; it copies only this skill's folder. If the author's install steps above differ, follow those first. To scope it to one project, replace ~/.claude/skills with that project's .claude/skills.
How do you use this skill?
With the skill installed, give the coding agent a request such as: Create or update an Azure SDK skill for <SDK package>, using <documentation URL or repository reference>. Supply the required package name, documentation URL, or repository reference. The supplied material does not fully document the skill's runtime trigger mechanism or invocation command.
What are this skill's strengths and limitations?
- Defines a concrete folder structure, section order, and frontmatter contract.
- Provides detailed Azure SDK guidance for token budgets, authentication, credential selection, and client lifecycle.
- Includes anti-pattern checklists and optional validation workflows for quality control.
- The repository is MIT-licensed and described as actively maintained, with documentation and tests updated through workflows.
- The guidance is strongly oriented toward Azure SDKs and Microsoft Foundry; authors of unrelated domain skills must organize those skills themselves.
- SDK/API skill creation requires a package name, documentation URL, or repository reference before work can begin.
- Some validation paths depend on microsoft-docs MCP, Vally, or a test environment; universal availability is not established by the source.
- The supplied SKILL.md is explicitly truncated, so its unseen portions remain unknown.
How does this skill compare with similar options?
Side by side with related skills; every score comes from the same FSRS standard.
| Skill | FS score | Stars | Last updated | License |
|---|---|---|---|---|
| Azure Skill Creator this page ✓ Microsoft · Official | 48 · Use with care | ★ 2.7k | 3mo ago | MIT |
| Microsoft Agent Skills ✓ Microsoft · Official | Not yet reviewed | ★ 2.7k | 3mo ago | MIT |
| Microsoft Foundry Agent Skill ✓ Microsoft · Official | 47 · Use with care | ★ 2.7k | 3mo ago | MIT |
| Azure AI Foundry Agent Framework ✓ Microsoft · Official | 50 · Use with care | ★ 2.7k | 3mo ago | MIT |
| Azure AI Projects for TypeScript ✓ Microsoft · Official | 49 · Use with care | ★ 2.7k | 3mo ago | MIT |
How did FollowSkills review this skill?
The guidance reduces credential risk through prohibition of hardcoded secrets, constrained DefaultAzureCredential usage, and references to official documentation; however, it does not define user confirmation, least-privilege permissions, data-flow disclosure, approval for external effects, or rollback, so points are deducted.
The document is fairly internally structured and supplies initialization, validation, and packaging scripts with failure messages; however, no tests are provided, dependency availability and key-path reproduction are unverified, edge-case coverage is thin, and some examples may not be directly runnable, so the score remains conservatively limited.
The audience and main scenarios—creating or updating skills for Azure SDKs and Microsoft Foundry—are reasonably clear; boundaries for non-Azure skills, agent environments, Chinese-language use, and mainland-China network reachability are under-specified, as are non-trigger and false-trigger conditions.
It includes frontmatter, progressive-disclosure guidance, directory conventions, references, script documentation, naming rules, and validation guidance. The supplied MIT metadata and verified Microsoft provenance support governance, but the skill lacks its own versioning, changelog, named maintenance responsibility, and explicit update path, while relying on many prescriptive assumptions, so points are deducted.
The skill offers useful principles, Azure SDK authentication patterns, structural templates, and basic tooling; however, the initializer creates mostly placeholder content, the validator performs only basic format checks, and no end-to-end evidence shows production-ready skill output, so the score stays below the static ceiling.
The material cites official Azure SDK sources and exposes auditable scripts and CI/evaluation commands; nevertheless, the selected skill has no committed test suite, execution results, or independent evidence covering its key paths, so the score is limited.
Open a dimension to read why it scored that way
Evidence confidence：Low — Mostly static review, author material or a limited demo; useful for discovery, not high-risk decisions.
See the full review method →