skill-builder
について
このClaudeスキルは、Anthropicの公式ベストプラクティスとamplihack哲学に従って、開発者が本番環境対応のClaude Codeスキルを作成、改良、検証するのを支援します。新しいスキルの構築、作成、設計について言及すると自動的に起動します。このスキルは、要件を明確化し構造を設計する専門エージェントを使用して、開発プロセスを統括します。
クイックインストール
Claude Code
推奨/plugin add https://github.com/rysweet/MicrosoftHackathon2025-AgenticCodinggit clone https://github.com/rysweet/MicrosoftHackathon2025-AgenticCoding.git ~/.claude/skills/skill-builderこのコマンドをClaude Codeにコピー&ペーストしてスキルをインストールします
ドキュメント
Skill Builder
Purpose
Helps users create production-ready Claude Code skills that follow best practices from official Anthropic documentation and amplihack's ruthless simplicity philosophy.
When I Activate
I automatically load when you mention:
- "build a skill" or "create a skill"
- "generate a skill" or "make a skill"
- "design a skill" or "develop a skill"
- "skill builder" or "new skill"
- "skill for [purpose]"
What I Do
I orchestrate the skill creation process using amplihack's specialized agents:
-
Clarify Requirements (prompt-writer agent)
- Understand skill purpose and scope
- Define target users and use cases
- Identify skill type (agent, command, scenario)
-
Design Structure (architect agent)
- Plan YAML frontmatter fields
- Design skill organization (single vs multi-file)
- Calculate token budget allocation
- Choose appropriate templates
-
Generate Skill (builder agent)
- Create SKILL.md with proper YAML frontmatter
- Write clear instructions and examples
- Include supporting files if needed
- Follow progressive disclosure pattern
-
Validate Quality (reviewer agent)
- Check YAML frontmatter syntax
- Verify token budget (<5,000 tokens core)
- Ensure philosophy compliance (>85% score)
- Test description quality for discovery
-
Create Tests (tester agent)
- Define activation test cases
- Create edge case validations
- Document expected behaviors
Skill Types Supported
- skill: Claude Code skills in
.claude/skills/(auto-discovery) - agent: Specialized agents in
.claude/agents/amplihack/specialized/ - command: Slash commands in
.claude/commands/amplihack/ - scenario: Production tools in
.claude/scenarios/
See examples.md for detailed examples of each type.
Command Interface
For explicit invocation:
/amplihack:skill-builder <skill-name> <skill-type> <description>
Examples in examples.md.
Documentation
Supporting Files (progressive disclosure):
- reference.md: Architecture, patterns, YAML spec, best practices
- examples.md: Real-world usage, testing, troubleshooting
Original Documentation Sources (embedded in reference.md):
- Official Claude Code Skills: https://code.claude.com/docs/en/skills
- Anthropic Agent SDK Skills: https://docs.claude.com/en/docs/agent-sdk/skills
- Agent Skills Engineering Blog: https://www.anthropic.com/engineering/equipping-agents-for-the-real-world-with-agent-skills
- Claude Cookbooks - Skills: https://github.com/anthropics/claude-cookbooks/tree/main/skills
- Skills Custom Development Notebook: https://github.com/anthropics/claude-cookbooks/blob/main/skills/notebooks/03_skills_custom_development.ipynb
- metaskills/skill-builder (Reference): https://github.com/metaskills/skill-builder
All documentation is embedded in reference.md for offline access. Links provided for updates and verification.
Note: This skill automatically loads when Claude detects skill building intent. For explicit control, use /amplihack:skill-builder.
GitHub リポジトリ
関連スキル
content-collections
メタThis skill provides a production-tested setup for Content Collections, a TypeScript-first tool that transforms Markdown/MDX files into type-safe data collections with Zod validation. Use it when building blogs, documentation sites, or content-heavy Vite + React applications to ensure type safety and automatic content validation. It covers everything from Vite plugin configuration and MDX compilation to deployment optimization and schema validation.
creating-opencode-plugins
メタThis skill provides the structure and API specifications for creating OpenCode plugins that hook into 25+ event types like commands, files, and LSP operations. It offers implementation patterns for JavaScript/TypeScript modules that intercept and extend the AI assistant's lifecycle. Use it when you need to build event-driven plugins for monitoring, custom handling, or extending OpenCode's capabilities.
evaluating-llms-harness
テストThis Claude Skill runs the lm-evaluation-harness to benchmark LLMs across 60+ standardized academic tasks like MMLU and GSM8K. It's designed for developers to compare model quality, track training progress, or report academic results. The tool supports various backends including HuggingFace and vLLM models.
sglang
メタSGLang is a high-performance LLM serving framework that specializes in fast, structured generation for JSON, regex, and agentic workflows using its RadixAttention prefix caching. It delivers significantly faster inference, especially for tasks with repeated prefixes, making it ideal for complex, structured outputs and multi-turn conversations. Choose SGLang over alternatives like vLLM when you need constrained decoding or are building applications with extensive prefix sharing.
