Skip to main content Inicio Creadores ajbcoding claude-skill-eval moai-alfred-rules
moai-alfred-rules Alfred SuperAgent의 필수 규칙을 정의합니다. November 2025 enterprise standard 기반. 3-Layer architecture, 4-Step workflow, Agent-first paradigm, Skill invocation rules, AskUserQuestion patterns, TRUST 5 quality gates, TAG chain integrity, commit message standards. 사용: 워크플로우 규칙 검증, 품질 게이트 확인, MoAI-ADK 표준 준수, 아키텍처 규칙 검증.
Ir a la instalación Skills Marketplace Descubre y explora habilidades de IA creadas por la comunidad.
Instalar con Codex o Claude Copia este prompt, pégalo en Codex, Claude u otro asistente, y deja que revise la página de la skill y la instale por ti.
Mostrar detalles del prompt Un comando directo omite el prompt de revisión. Revisa el origen antes de ejecutarlo.
npx skills add https://github.com/AJBcoding/claude-skill-eval --skill moai-alfred-rulesEl comando permanece en una sola línea. Desplázate horizontalmente para revisarlo antes de copiarlo.
¿Prefieres una copia local? Descarga los archivos que SkillsMP tiene disponibles ahora.
Descargar Zip Descargando... Más de este repositorio Enterprise database architecture specialist with PostgreSQL 17, MySQL 8.4 LTS,
MongoDB 8.0, Redis 7.4 expertise. Master connection pooling, query optimization,
caching strategies, and database DevOps automation. Build scalable, resilient
database systems with comprehensive monitoring and disaster recovery.
Enterprise Frontend Development with AI-powered modern architecture, Context7 integration, and intelligent component orchestration for scalable user interfaces
Enterprise-grade security expertise with production-ready patterns for OWASP Top 10 2021, zero-trust architecture, threat modeling (STRIDE, PASTA), secure SDLC, DevSecOps automation, cloud security, cryptography, identity & access management, and compliance frameworks (SOC 2, ISO 27001, GDPR, CCPA).
Ocupaciones relacionadas SOC
Basado en la clasificación ocupacional SOC
Explorador de archivos
3 archivos name moai-alfred-rules version 4.0.0 created 2025-11-02T00:00:00.000Z updated 2025-11-12T00:00:00.000Z status stable tier Alfred description Alfred SuperAgent의 필수 규칙을 정의합니다. November 2025 enterprise standard 기반. 3-Layer architecture, 4-Step workflow, Agent-first paradigm, Skill invocation rules, AskUserQuestion patterns, TRUST 5 quality gates, TAG chain integrity, commit message standards. 사용: 워크플로우 규칙 검증, 품질 게이트 확인, MoAI-ADK 표준 준수, 아키텍처 규칙 검증. keywords ["rules","agent-first","skill-invocation","ask-user-question","trust-5","tag-chain","workflow-compliance","quality-gates","4-step-workflow","architecture-rules"] allowed-tools Read, Glob, Grep, Bash, AskUserQuestion, TodoWrite
Skill 개요
moai-alfred-rules 는 Alfred SuperAgent의 의사결정과 실행을 제어하는 핵심 프레임워크입니다.
항목 값 버전 4.0.0 (November 2025 enterprise) 티어 Alfred (상위 계층) 자동 로드 규칙 검증, 품질 게이트, 아키텍처 규칙 필요 시 아키텍처 패러다임 Agent-First (Command → Agent → Skill → Hook) 워크플로우 모델 4-Step ADAP Workflow
무엇을 하는가?
핵심 책임
3-Layer Architecture 정의 : Commands → Agents → Skills 계층 분리4-Step Workflow 규칙 : ADAP (Analyze, Design, Assure, Produce) + IntentAgent-First Paradigm : 모든 실행 작업을 agents에 위임Skill 호출 규칙 : 10+ mandatory patterns, invocation syntaxAskUserQuestion 패턴 : 5가지 필수 사용 시나리오TRUST 5 Quality Gates : T/R/U/S/T 각각의 검증 기준TAG Chain Integrity : SPEC→TEST→CODE→DOC 추적Commit Message Standards : TDD cycle message formats
언제 사용하는가?
필수 시나리오 (MUST use)
상황 사용 여부 ✅ Skill() 호출 규칙 검증 필수 ✅ Command vs Agent 역할 분리 필수 ✅ AskUserQuestion 사용 판단 필수 ✅ TRUST 5 준수 확인 필수 ✅ TAG 체인 무결성 검증 필수 ✅ 커밋 메시지 형식 확인 필수 ✅ 워크플로우 compliance 검증 필수 ✅ Agent delegation 올바른지 확인 필수 ✅ 품질 게이트(quality gate) 통과 필수 ✅ 아키텍처 규칙 위반 감지 필수
Rule 1: 3-Layer Architecture (November 2025 Standard)
계층 구조
┌─────────────────────────────────────┐
│ Commands (Orchestration Only) │ ← User-facing entry points
│ /alfred:0-project, /alfred:1-plan │ No direct execution
│ /alfred:2-run, /alfred:3-sync │
└──────────┬──────────────────────────┘
│ Task(subagent_type="...")
↓
┌─────────────────────────────────────┐
│ Agents (Domain Expertise) │ ← Deep reasoning
│ spec-builder, tdd-implementer │ Complex decisions
│ test-engineer, doc-syncer │ Plan → Execute
│ git-manager, qa-validator │
└──────────┬──────────────────────────┘
│ Skill("skill-name")
↓
┌─────────────────────────────────────┐
│ Skills (Knowledge Capsules) │ ← Reusable patterns
│ 55 specialized Skills │ Playbooks
│ < 1000 lines each │ Best practices
└─────────────────────────────────────┘
규칙 1.1: Commands - Orchestration ONLY
echo "Building application..."
python setup.py build
git commit -m "Build"
Skill("moai-alfred-rules" )
if feature_type == "backend" :
Task(
subagent_type="tdd-implementer" ,
description="Build and test application" ,
prompt="Implement feature with RED-GREEN-REFACTOR cycle"
)
if user_approval_needed:
AskUserQuestion(...)
Task(subagent_type="implementation-planner" , ...)
규칙 1.2: Agents - Domain Expertise Ownership
✅ 복잡한 분석 & 추론 (deep reasoning)
✅ 계획 수립 (planning)
✅ 의사결정 (decision-making)
✅ Skill 호출 및 조율 (orchestration within domain)
✅ 작업 실행 (execution)
예시: tdd-implementer Agent :
Agent receives: "Implement user authentication"
↓
1. Analyze: SPEC 검토, 요구사항 분석
2. Design: 아키텍처 설계
3. RED: test-engineer Skill 호출 → 테스트 작성
4. GREEN: 코드 구현
5. REFACTOR: 최적화
6. Commit: git-manager 호출 → commit
↓
Returns: Fully tested, documented, committed code
규칙 1.3: Skills - Knowledge Capsules (Stateless)
✅ 상태가 없음 (stateless)
✅ 재사용 가능 (reusable)
✅ Agent에 의해 호출됨 (called by agents)
✅ 1000줄 이하 (< 1000 lines)
✅ 단일 주제 (single topic)
Skill("moai-foundation-git" )
Task(subagent_type="..." )
state = {"counter" : 0}
Rule 2: 4-Step Agent-Based Workflow (November 2025)
Phase Overview Phase 0: INTENT Phase 1: ANALYZE
┌──────────────────┐ ┌──────────────────┐
│ User Request │ ─clarity?─→ │ WebSearch │
│ Ambiguous? │ NO │ WebFetch │
│ ✅ YES: clarify │ YES │ Research │
│ ✅ NO: continue │────────────→ │ Best practices │
└──────────────────┘ └──────────────────┘
↓
Phase 3: ASSURE Phase 2: DESIGN
┌──────────────────┐ ┌──────────────────┐
│ Quality Gate │←──────────────→│ Architecture │
│ TRUST 5 │ │ Latest info │
│ TAG integrity │ │ Version specs │
│ Compliance │ │ Design patterns │
└──────────────────┘ └──────────────────┘
↓
Phase 4: PRODUCE
┌──────────────────┐
│ Skill invocation │
│ File generation │
│ Commit │
└──────────────────┘
Phase 0: Intent (사용자 의도 파악) 규칙 0.1 : Intent가 모호하면 AskUserQuestion 사용
상황: "데이터 처리 모듈 만들어줘"
Step 1: Intent 평가
├─ Clarity: LOW (어떤 데이터? 어떤 처리?)
└─ Action: AskUserQuestion 사용
Step 2: 명확화
AskUserQuestion({
question: "어떤 데이터를 처리하나요?",
options: [
"CSV 파일",
"데이터베이스",
"API 응답"
]
})
Step 3: 명확한 요구사항 확보 후 다음 phase로
Phase 1: Analyze (정보 수집 & 연구) Task 1: Research latest information
├─ Search: "[framework] [version] best practices 2025"
├─ Fetch: Official documentation URLs
└─ Validate: Cross-check multiple sources
Task 2: Collect best practices
├─ Official docs
├─ Industry standards
└─ Current patterns (2025)
Task 3: Identify version-specific guidance
├─ Current stable version
├─ Breaking changes
└─ Deprecation warnings
Phase 2: Design (아키텍처 설계) 입력 : Phase 1의 연구 결과
출력 : November 2025 최신 정보를 기반한 설계
Design Activities:
├─ 최신 정보 기반 이름 지정
├─ 현재 버전 명시
├─ 최신 패턴 포함
├─ 공식 문서 링크
└─ 마지막 업데이트 날짜 포함
Phase 3: Assure (품질 검증) Gate 검증 기준 Test 85%+ coverage, 모든 경로 테스트 Readable Clean code, SOLID 원칙, 주석 포함 Unified 일관된 패턴, 중복 제거, 네이밍 표준 Secured OWASP Top 10 확인, 비밀정보 제거
Phase 4: Produce (생성 & 배포) 책임 : Skill 호출 (예: moai-skill-factory)
Actions:
├─ 템플릿 적용
├─ 파일 생성
├─ 메타데이터 추가
├─ 최신 정보 embedded
├─ 공식 문서 링크 포함
└─ Version date 표기
Rule 3: Agent-First Paradigm (Critical Enforcement)
금지된 작업 (❌ ABSOLUTELY FORBIDDEN) Alfred (또는 Command)가 직접 실행하면 절대 안 됩니다:
직접 bash 명령 실행 ❌
bash("git commit -m 'message'" )
os.system("python build.py" )
Task(subagent_type="git-manager" , prompt="Commit changes" )
파일 읽기/쓰기 ❌
with open("file.py" , "w" ) as f:
f.write(code)
Task(subagent_type="file-manager" , prompt="Create file" )
Git 직접 조작 ❌
subprocess.run(["git" , "push" , "origin" , "main" ])
Task(subagent_type="git-manager" , prompt="Push changes" )
코드 분석 직접 실행 ❌
lines = len(open("file.py" ).readlines())
Task(subagent_type="code-analyzer" , prompt="Analyze code" )
테스트 직접 실행 ❌
subprocess.run(["pytest" , "tests/" ])
Task(subagent_type="test-engineer" , prompt="Run tests" )
의무 위임 (✅ MANDATORY DELEGATION) 작업 위임 대상 패턴 계획 수립 plan-agent Task(subagent_type="plan-agent", ...)코드 개발 tdd-implementer Task(subagent_type="tdd-implementer", ...)테스트 작성 test-engineer Task(subagent_type="test-engineer", ...)문서화 doc-syncer Task(subagent_type="doc-syncer", ...)Git 작업 git-manager Task(subagent_type="git-manager", ...)품질 검증 qa-validator Task(subagent_type="qa-validator", ...)사용자 질문 ask-user-questions AskUserQuestion(...)
Rule 4: 10 Mandatory Skill Invocations
규칙 4.1: Skill Invocation Pattern
10가지 필수 Skill # Skill 용도 Invocation 1 moai-foundation-trust TRUST 5 검증 Skill("moai-foundation-trust")2 moai-foundation-tags TAG 검증 & 추적 Skill("moai-foundation-tags")3 moai-foundation-specs SPEC 작성 & 검증 Skill("moai-foundation-specs")4 moai-foundation-ears EARS 요구사항 형식 Skill("moai-foundation-ears")5 moai-foundation-git Git 워크플로우 Skill("moai-foundation-git")6 moai-foundation-langs 언어 & 스택 감지 Skill("moai-foundation-langs")7 moai-essentials-debug 디버깅 & 에러 분석 Skill("moai-essentials-debug")8 moai-essentials-refactor 리팩토링 & 개선 Skill("moai-essentials-refactor")9 moai-essentials-perf 성능 최적화 Skill("moai-essentials-perf")10 moai-essentials-review 코드 리뷰 & 품질 Skill("moai-essentials-review")
규칙 4.2: Skill Invocation 예제
if validation_required:
Skill("moai-foundation-trust" )
if tag_integrity_check:
Skill("moai-foundation-tags" )
if spec_needed:
Skill("moai-foundation-specs" )
if git_decision_needed:
Skill("moai-foundation-git" )
Rule 5: AskUserQuestion Patterns (5가지 필수 시나리오)
규칙 5.1: MANDATORY Scenarios 상황: "Python 웹 프레임워크 추천해줄래?"
AskUserQuestion({
question: "어떤 유형의 애플리케이션?",
header: "Application Type",
options: [
{ label: "REST API", description: "High performance APIs" },
{ label: "Web Application", description: "Traditional MVC" },
{ label: "Microservice", description: "Event-driven" }
]
})
상황: "데이터베이스 모델을 어떻게 설계?"
AskUserQuestion({
question: "어떤 데이터 특성?",
header: "Data Model",
options: [
{ label: "Relational", description: "Structured, ACID" },
{ label: "Document", description: "Flexible schema" },
{ label: "Graph", description: "Relationships" }
]
})
상황: "코드 개선해줄래?"
AskUserQuestion({
question: "어떤 측면을 개선?",
header: "Improvement Focus",
options: [
{ label: "Performance", description: "Speed & efficiency" },
{ label: "Readability", description: "Code clarity" },
{ label: "Security", description: "Vulnerability fixes" }
],
multiSelect: true # 복수 선택 허용
})
상황: "패키지 업그레이드하려는데 호환성?"
AskUserQuestion({
question: "기존 코드 호환성 유지 필요?",
header: "Compatibility",
options: [
{ label: "Full compatibility", description: "Maintain all APIs" },
{ label: "Deprecation path", description: "Gradual migration" },
{ label: "Breaking OK", description: "Version bump allowed" }
]
})
상황: "시스템 리팩토링 계획 수립"
AskUserQuestion({
question: "예상 개발 기간?",
header: "Timeline",
options: [
{ label: "1 week", description: "Focused scope" },
{ label: "2-4 weeks", description: "Medium refactor" },
{ label: "1+ months", description: "Comprehensive" }
]
})
규칙 5.2: 올바른 사용법 사용자: "뭘 선호해?"
응답: "음... 생각해봤는데 아마도..."
✅ CORRECT (AskUserQuestion):
AskUserQuestion({
question: "어떤 접근을 선호하시나요?",
header: "Approach",
multiSelect: false,
options: [
{ label: "Option A", description: "Benefit A, Cost B" },
{ label: "Option B", description: "Benefit C, Cost D" }
]
})
Rule 6: TRUST 5 Quality Gates (November 2025 Enterprise)
각 Gate의 검증 기준
T: Test First (85%+ Coverage) requirements:
coverage: "≥ 85%"
coverage_tools: ["pytest-cov" , "coverage.py" ]
test_types:
- unit_tests: "각 함수/메서드"
- integration_tests: "모듈 간 상호작용"
- edge_cases: "경계값, 에러 처리"
validation:
- All code paths executed
- Error conditions tested
- Mock external dependencies
- No skipped tests (×skip, ×xfail)
R: Readable (Clean Code) requirements:
code_standards:
- SOLID 원칙 준수
- DRY (Don't Repeat Yourself)
- KISS (Keep It Simple, Stupid)
documentation:
- Function docstrings
- Complex logic comments
- Type hints (Python 3.10 +)
naming:
- Descriptive variable names
- Consistent conventions
- No single-letter vars (except i, j, k in loops)
U: Unified (Consistent Patterns) requirements:
consistency:
- Same patterns across codebase
- No duplicate logic
- Shared utilities for common tasks
conventions:
- Consistent indentation (4 spaces)
- Consistent naming (snake_case, PascalCase)
- Consistent import order
validation:
- Linting (flake8, black, isort)
- Static analysis (pylint, mypy)
- No code duplication (DRY violations)
S: Secured (OWASP Top 10) requirements:
security_checks:
- No hardcoded credentials
- No SQL injection vectors
- No XXE vulnerabilities
- Input validation
- Output encoding
tools:
- bandit (Python security linter)
- safety (dependency vulnerabilities)
- SAST scanning
validation:
- No secrets committed
- Dependencies scanned
- Known CVEs checked
requirements:
tag_chain:
- Complete history traceability
documentation:
validation:
- Bidirectional references
Rule 7: TAG Chain Integrity Rules
규칙 7.1: TAG Naming Convention Format: @<DOMAIN>-<###>
Examples:
규칙 7.2: TAG Lifecycle
└─ Example: test_auth.py - test_auth_success()
└─ Example: auth.py - authenticate_user()
└─ Example: CHANGELOG.md
규칙 7.3: TAG Validation Rules
Rule 8: Commit Message Standards (TDD Cycle)
규칙 8.1: Commit Format Format:
<type>(<tag>): <subject>
<body>
<footer>
규칙 8.2: TDD Cycle Commits
- test_successful_login()
- test_invalid_credentials()
- test_expired_token()
Status: RED (Tests fail as expected)
- Implement authenticate_user() function
- Add token generation
- Add error handling
Status: GREEN (All tests pass)
- Extract token validation to separate function
- Add caching for user lookups
- Improve error messages
Status: PASSING (Tests still pass, code improved)
규칙 8.3: Commit 타입 (Conventional Commits) 타입 설명 예시 chore 빌드/설정 chore: Update dependencies
Rule 9: Workflow Compliance Validation
규칙 9.1: Compliance Checklist
규칙 9.2: Violation Response 1. 자동 탐지 (hook)
2. 사용자에게 알림
3. 수정 요청
4. 재검증
5. Pass/Fail 결정
3-Level Progressive Disclosure
Level 1: 빠른 시작 (Beginner - 10분)
Command = 조율만, Agent = 실행
Skill = 전문 도구
규칙 위반 → 에러
Level 2: 실무 패턴 (Intermediate - 30분)
AskUserQuestion 사용 시점 판단
TRUST 5 검증
TAG 체인 유지
커밋 메시지 형식
Level 3: 심화 (Advanced - 1시간)
Agent delegation 전략
Skill 조합 활용
Workflow 자동화
규칙 예외 관리
공식 자료 & 참고 링크
Architecture References
Command-Agent-Skill 패턴: Internal moai-adk documentation
ADAP Workflow: Internal workflow definition
Quality Standards
TRUST 5 Framework: Skill("moai-foundation-trust")
TAG System: Skill("moai-foundation-tags")
Commit Standards: Skill("moai-foundation-git")
Enterprise Standards (November 2025)
Agent-First Architecture: InfoQ (agentic-ai-architecture-framework)
TDD Best Practices: Official pytest documentation
Code Quality: OWASP Top 10, SOLID principles
버전 : 4.0.0 (November 2025 Enterprise Standard)
최종 업데이트 : 2025-11-12
유지보수자 : GoosLab (Alfred SuperAgent Framework)
