moai-adk 0.4.0__py3-none-any.whl → 0.4.4__py3-none-any.whl

moai_adk/templates/.claude/skills/moai-domain-security/SKILL.md

@@ -1,23 +1,32 @@
 ---
+
 name: moai-domain-security
-description: OWASP Top 10, static analysis (SAST), dependency security, and secrets
-  management
+description: OWASP Top 10, static analysis (SAST), dependency security, and secrets management. Use when working on security and compliance reviews scenarios.
 allowed-tools:
-- Read
-- Bash
+  - Read
+  - Bash
 ---
 
 # Security Expert
 
+## Skill Metadata
+| Field | Value |
+| ----- | ----- |
+| Allowed tools | Read (read_file), Bash (terminal) |
+| Auto-load | On demand when security keywords appear |
+| Trigger cues | Threat modeling, OWASP findings, secrets management, compliance reviews. |
+| Tier | 4 |
+
 ## What it does
 
 Provides expertise in application security, including OWASP Top 10 vulnerabilities, static application security testing (SAST), dependency vulnerability scanning, and secrets management.
 
 ## When to use
 
-- "보안 취약점 분석", "OWASP 검증", "시크릿 관리", "의존성 보안"
+- Engages when the team asks about security posture or mitigation steps.
+- “Security vulnerability analysis”, “OWASP verification”, “Secret management”, “Dependency security”
 - Automatically invoked when security concerns arise
-- Security SPEC implementation (`/alfred:2-build`)
+- Security SPEC implementation (`/alfred:2-run`)
 
 ## How it works
 
@@ -58,17 +67,39 @@ Provides expertise in application security, including OWASP Top 10 vulnerabiliti
 - HTTPS enforcement
 
 ## Examples
+```markdown
+- Run SAST/DAST tools and attach findings summary.
+- Update risk matrix with severity/owner/ETA.
+```
+
+## Inputs
+- Domain-related design documents and user requirements.
+- Project technology stack and operational constraints.
+
+## Outputs
+- Domain-specific architecture or implementation guidelines.
+- Recommended list of associated sub-agents/skills.
 
-### Example 1: OWASP compliance check
-User: "/alfred:2-build SEC-001"
-Claude: (creates RED security test, GREEN implementation with input validation, REFACTOR)
+## Failure Modes
+- When the domain document does not exist or is ambiguous.
+- When the project strategy is unconfirmed and cannot be specified.
 
-### Example 2: Dependency vulnerability scan
-User: "의존성 보안 스캔"
-Claude: (runs npm audit or snyk test and reports vulnerabilities)
+## Dependencies
+- `.moai/project/` document and latest technical briefing are required.
+
+## References
+- OWASP. "Top 10 Web Application Security Risks." https://owasp.org/www-project-top-ten/ (accessed 2025-03-29).
+- NIST. "Secure Software Development Framework." https://csrc.nist.gov/publications/detail/sp/800-218/final (accessed 2025-03-29).
+
+## Changelog
+- 2025-03-29: Codified input/output and failure responses for domain skills.
 
 ## Works well with
 
 - alfred-trust-validation (security validation)
 - web-api-expert (API security)
 - devops-expert (secure deployments)
+
+## Best Practices
+- Record supporting documentation (version/link) for each domain decision.
+- Review performance, security, and operational requirements simultaneously at an early stage.

moai_adk/templates/.claude/skills/moai-domain-web-api/SKILL.md

@@ -1,23 +1,32 @@
 ---
+
 name: moai-domain-web-api
-description: REST API and GraphQL design patterns with authentication, versioning,
-  and OpenAPI documentation
+description: REST API and GraphQL design patterns with authentication, versioning, and OpenAPI documentation. Use when working on web API contracts scenarios.
 allowed-tools:
-- Read
-- Bash
+  - Read
+  - Bash
 ---
 
 # Web API Expert
 
+## Skill Metadata
+| Field | Value |
+| ----- | ----- |
+| Allowed tools | Read (read_file), Bash (terminal) |
+| Auto-load | On demand for API delivery |
+| Trigger cues | REST/GraphQL design, contract testing, versioning, integration hardening. |
+| Tier | 4 |
+
 ## What it does
 
 Provides expertise in designing and implementing RESTful APIs and GraphQL services, including authentication mechanisms (JWT, OAuth2), API versioning strategies, and OpenAPI documentation.
 
 ## When to use
 
-- "API 설계", "REST API 패턴", "GraphQL 스키마", "JWT 인증"
+- Engages when designing or validating web APIs and their lifecycle controls.
+- “API design”, “REST API pattern”, “GraphQL schema”, “JWT authentication”
 - Automatically invoked when working with API projects
-- Web API SPEC implementation (`/alfred:2-build`)
+- Web API SPEC implementation (`/alfred:2-run`)
 
 ## How it works
 
@@ -50,17 +59,39 @@ Provides expertise in designing and implementing RESTful APIs and GraphQL servic
 - **Postman collections**: Request examples
 
 ## Examples
+```bash
+$ uvicorn app.main:app --reload
+$ newman run postman_collection.json
+```
+
+## Inputs
+- Domain-related design documents and user requirements.
+- Project technology stack and operational constraints.
+
+## Outputs
+- Domain-specific architecture or implementation guidelines.
+- Recommended list of associated sub-agents/skills.
 
-### Example 1: REST API with JWT
-User: "/alfred:2-build API-AUTH-001"
-Claude: (creates RED API test, GREEN implementation with JWT middleware, REFACTOR)
+## Failure Modes
+- When the domain document does not exist or is ambiguous.
+- When the project strategy is unconfirmed and cannot be specified.
 
-### Example 2: GraphQL schema design
-User: "GraphQL 스키마 설계"
-Claude: (designs schema with proper types and resolvers)
+## Dependencies
+- `.moai/project/` document and latest technical briefing are required.
+
+## References
+- Microsoft. "REST API Design Guidelines." https://learn.microsoft.com/azure/architecture/best-practices/api-design (accessed 2025-03-29).
+- OpenAPI Initiative. "OpenAPI Specification." https://spec.openapis.org/oas/latest.html (accessed 2025-03-29).
+
+## Changelog
+- 2025-03-29: Codified input/output and failure responses for domain skills.
 
 ## Works well with
 
 - alfred-trust-validation (API security validation)
 - backend-expert (server implementation)
 - security-expert (authentication patterns)
+
+## Best Practices
+- Record supporting documentation (version/link) for each domain decision.
+- Review performance, security, and operational requirements simultaneously at an early stage.

moai_adk/templates/.claude/skills/moai-essentials-debug/SKILL.md

@@ -1,26 +1,34 @@
 ---
+
 name: moai-essentials-debug
-description: Advanced debugging with stack trace analysis, error pattern detection,
-  and fix suggestions
+description: Advanced debugging with stack trace analysis, error pattern detection, and fix suggestions. Use when delivering quick diagnostic support for everyday issues.
 allowed-tools:
-- Read
-- Bash
-- Write
-- Edit
-- TodoWrite
+  - Read
+  - Bash
+  - Write
+  - Edit
+  - TodoWrite
 ---
 
 # Alfred Debugger Pro
 
+## Skill Metadata
+| Field | Value |
+| ----- | ----- |
+| Allowed tools | Read (read_file), Write (write_file), Edit (edit_file), Bash (terminal), TodoWrite (todo_write) |
+| Auto-load | On demand during Run stage (debug-helper) |
+| Trigger cues | Runtime error triage, stack trace analysis, root cause investigation requests. |
+
 ## What it does
 
 Advanced debugging support with stack trace analysis, common error pattern detection, and actionable fix suggestions.
 
 ## When to use
 
-- "에러 해결해줘", "이 오류 원인은?", "스택 트레이스 분석"
+- Loads when users share stack traces or ask why a failure occurred.
+- “Resolve the error”, “What is the cause of this error?”, “Stack trace analysis”
 - Automatically invoked on runtime errors (via debug-helper sub-agent)
-- "왜 안 돼?", "NullPointerException 해결"
+- "Why not?", "Solving NullPointerException"
 
 ## How it works
 
@@ -58,9 +66,37 @@ jwt.exceptions.ExpiredSignatureError: Signature has expired
 - **Java**: Optional, try-with-resources
 
 ## Examples
+```markdown
+- Checks the current diff and lists items that can be modified immediately.
+- Schedule follow-up tasks with TodoWrite.
+```
+
+## Inputs
+- A snapshot of the code/tests/documentation you are currently working on.
+- Ongoing agent status information.
+
+## Outputs
+- Immediately actionable checklists or improvement suggestions.
+- Recommendations on whether to take next steps or not.
+
+## Failure Modes
+- If you cannot find the required files or test results.
+- When the scope of work is excessively large and cannot be resolved with simple support.
+
+## Dependencies
+- Mainly used in conjunction with `tdd-implementer`, `quality-gate`, etc.
+
+## References
+- Microsoft. "Debugging Techniques." https://learn.microsoft.com/visualstudio/debugger/ (accessed 2025-03-29).
+- JetBrains. "Debugging Code." https://www.jetbrains.com/help/idea/debugging-code.html (accessed 2025-03-29).
+
+## Changelog
+- 2025-03-29: Overhauled input/output definitions for Essentials skills.
 
-User: "JWT ExpiredSignatureError 해결해줘"
-Claude: (analyzes stack trace, identifies root cause, suggests fix)
 ## Works well with
 
 - moai-essentials-refactor
+
+## Best Practices
+- Record results, even for simple improvements, to increase traceability.
+- Clearly mark items that require human review to distinguish them from automation.

moai_adk/templates/.claude/skills/moai-essentials-perf/SKILL.md

@@ -1,25 +1,33 @@
 ---
+
 name: moai-essentials-perf
-description: Performance optimization with profiling, bottleneck detection, and tuning
-  strategies
+description: Performance optimization with profiling, bottleneck detection, and tuning strategies. Use when performing baseline performance reviews.
 allowed-tools:
-- Read
-- Bash
-- Write
-- Edit
-- TodoWrite
+  - Read
+  - Bash
+  - Write
+  - Edit
+  - TodoWrite
 ---
 
 # Alfred Performance Optimizer
 
+## Skill Metadata
+| Field | Value |
+| ----- | ----- |
+| Allowed tools | Read (read_file), Write (write_file), Edit (edit_file), Bash (terminal), TodoWrite (todo_write) |
+| Auto-load | On demand during Run stage (performance triage) |
+| Trigger cues | Latency complaints, profiling requests, throughput bottleneck analysis. |
+
 ## What it does
 
 Performance analysis and optimization with profiling tools, bottleneck detection, and language-specific optimization techniques.
 
 ## When to use
 
-- "성능 개선해줘", "느린 부분 찾아줘", "최적화 방법은?"
-- "프로파일링", "병목 지점", "메모리 누수"
+- Loads when diagnosing performance regressions or planning optimization work.
+- “Improve performance”, “Find slow parts”, “How to optimize?”
+- “Profiling”, “Bottleneck”, “Memory leak”
 
 ## How it works
 
@@ -60,9 +68,37 @@ Performance analysis and optimization with profiling tools, bottleneck detection
 - CPU usage: <70%
 
 ## Examples
+```markdown
+- Checks the current diff and lists items that can be modified immediately.
+- Schedule follow-up tasks with TodoWrite.
+```
+
+## Inputs
+- A snapshot of the code/tests/documentation you are currently working on.
+- Ongoing agent status information.
+
+## Outputs
+- Immediately actionable checklists or improvement suggestions.
+- Recommendations on whether to take next steps or not.
+
+## Failure Modes
+- If you cannot find the required files or test results.
+- When the scope of work is excessively large and cannot be resolved with simple support.
+
+## Dependencies
+- Mainly used in conjunction with `tdd-implementer`, `quality-gate`, etc.
+
+## References
+- Google SRE. "The Four Golden Signals." https://sre.google/sre-book/monitoring-distributed-systems/ (accessed 2025-03-29).
+- Dynatrace. "Application Performance Monitoring Best Practices." https://www.dynatrace.com/resources/ebooks/ (accessed 2025-03-29).
+
+## Changelog
+- 2025-03-29: Overhauled input/output definitions for Essentials skills.
 
-User: "N+1 query 문제 해결해줘"
-Claude: (identifies N+1 queries, suggests eager loading, provides optimized code)
 ## Works well with
 
 - moai-essentials-refactor
+
+## Best Practices
+- Record results, even for simple improvements, to increase traceability.
+- Clearly mark items that require human review to distinguish them from automation.

moai_adk/templates/.claude/skills/moai-essentials-refactor/SKILL.md

@@ -1,32 +1,41 @@
 ---
+
 name: moai-essentials-refactor
-description: Refactoring guidance with design patterns and code improvement strategies
+description: Refactoring guidance with design patterns and code improvement strategies. Use when planning incremental refactors with safety nets.
 allowed-tools:
-- Read
-- Bash
-- Write
-- Edit
-- TodoWrite
+  - Read
+  - Bash
+  - Write
+  - Edit
+  - TodoWrite
 ---
 
 # Alfred Refactoring Coach
 
+## Skill Metadata
+| Field | Value |
+| ----- | ----- |
+| Allowed tools | Read (read_file), Write (write_file), Edit (edit_file), Bash (terminal), TodoWrite (todo_write) |
+| Auto-load | On demand during Run stage (refactor planning) |
+| Trigger cues | Refactoring plans, code smell cleanup, design pattern coaching. |
+
 ## What it does
 
 Refactoring guidance with design pattern recommendations, code smell detection, and step-by-step improvement plans.
 
 ## When to use
 
-- "리팩토링 도와줘", "이 코드 개선 방법은?", "디자인 패턴 적용"
-- "코드 정리", "중복 제거", "함수 분리"
+- Loads when the user asks how to restructure code or apply design patterns.
+- “Help with refactoring”, “How can I improve this code?”, “Apply design patterns” 
+- “Code organization”, “Remove duplication”, “Separate functions”
 
 ## How it works
 
 **Refactoring Techniques**:
-- **Extract Method**: 긴 메서드 분리
-- **Replace Conditional with Polymorphism**: 조건문 제거
-- **Introduce Parameter Object**: 매개변수 그룹화
-- **Extract Class**: 거대한 클래스 분리
+- **Extract Method**: Separate long methods
+- **Replace Conditional with Polymorphism**: Remove conditional statements
+- **Introduce Parameter Object**: Group parameters
+- **Extract Class**: Massive class separation
 
 **Design Pattern Recommendations**:
 - Complex object creation → **Builder Pattern**
@@ -51,9 +60,37 @@ Refactoring guidance with design pattern recommendations, code smell detection,
 - [ ] Commit frequently
 
 ## Examples
+```markdown
+- Checks the current diff and lists items that can be modified immediately.
+- Schedule follow-up tasks with TodoWrite.
+```
+
+## Inputs
+- A snapshot of the code/tests/documentation you are currently working on.
+- Ongoing agent status information.
+
+## Outputs
+- Immediately actionable checklists or improvement suggestions.
+- Recommendations on whether to take next steps or not.
+
+## Failure Modes
+- If you cannot find the required files or test results.
+- When the scope of work is excessively large and cannot be resolved with simple support.
+
+## Dependencies
+- Mainly used in conjunction with `tdd-implementer`, `quality-gate`, etc.
+
+## References
+- Fowler, Martin. "Refactoring: Improving the Design of Existing Code." Addison-Wesley, 2018.
+- IEEE Software. "Managing Technical Debt." IEEE Software, 2021.
+
+## Changelog
+- 2025-03-29: Overhauled input/output definitions for Essentials skills.
 
-User: "중복 코드 제거해줘"
-Claude: (identifies duplicates, suggests Extract Method, provides step-by-step plan)
 ## Works well with
 
 - moai-essentials-review
+
+## Best Practices
+- Record results, even for simple improvements, to increase traceability.
+- Clearly mark items that require human review to distinguish them from automation.

moai_adk/templates/.claude/skills/moai-essentials-review/SKILL.md

@@ -1,24 +1,32 @@
 ---
+
 name: moai-essentials-review
-description: Automated code review with SOLID principles, code smells, and language-specific
-  best practices
+description: Automated code review with SOLID principles, code smells, and language-specific best practices. Use when preparing concise review checklists for code changes.
 allowed-tools:
-- Read
-- Bash
-- Write
-- Edit
-- TodoWrite
+  - Read
+  - Bash
+  - Write
+  - Edit
+  - TodoWrite
 ---
 
 # Alfred Code Reviewer
 
+## Skill Metadata
+| Field | Value |
+| ----- | ----- |
+| Allowed tools | Read (read_file), Write (write_file), Edit (edit_file), Bash (terminal), TodoWrite (todo_write) |
+| Auto-load | On demand during Sync stage (review gate) |
+| Trigger cues | Code review requests, quality checklist preparation, merge readiness checks. |
+
 ## What it does
 
 Automated code review with language-specific best practices, SOLID principles verification, and code smell detection.
 
 ## When to use
 
-- "코드 리뷰해줘", "이 코드 개선점은?", "코드 품질 확인"
+- Loads when someone asks for a code review or a pre-merge quality assessment.
+- “Please review the code”, “How can this code be improved?”, “Check the code quality”
 - Optionally invoked after `/alfred:3-sync`
 - Before merging PR
 
@@ -67,10 +75,38 @@ Automated code review with language-specific best practices, SOLID principles ve
 ```
 
 ## Examples
+```markdown
+- Checks the current diff and lists items that can be modified immediately.
+- Schedule follow-up tasks with TodoWrite.
+```
+
+## Inputs
+- A snapshot of the code/tests/documentation you are currently working on.
+- Ongoing agent status information.
+
+## Outputs
+- Immediately actionable checklists or improvement suggestions.
+- Recommendations on whether to take next steps or not.
+
+## Failure Modes
+- If you cannot find the required files or test results.
+- When the scope of work is excessively large and cannot be resolved with simple support.
+
+## Dependencies
+- Mainly used in conjunction with `tdd-implementer`, `quality-gate`, etc.
+
+## References
+- IEEE. "Recommended Practice for Software Reviews." IEEE 1028-2008.
+- Cisco. "Peer Review Best Practices." https://www.cisco.com/c/en/us/support/docs/optical/ons-15454-esc/15114-peer-review.html (accessed 2025-03-29).
+
+## Changelog
+- 2025-03-29: Overhauled input/output definitions for Essentials skills.
 
-User: "이 코드 리뷰해줘"
-Claude: (analyzes code, detects issues, provides improvement suggestions)
 ## Works well with
 
 - moai-foundation-specs
 - moai-essentials-refactor
+
+## Best Practices
+- Record results, even for simple improvements, to increase traceability.
+- Clearly mark items that require human review to distinguish them from automation.

moai_adk/templates/.claude/skills/moai-foundation-ears/SKILL.md

@@ -1,23 +1,32 @@
 ---
+
 name: moai-foundation-ears
-description: EARS requirement authoring guide (Ubiquitous/Event/State/Optional/Constraints)
+description: EARS requirement authoring guide (Ubiquitous/Event/State/Optional/Constraints). Use when teams need guidance on EARS requirements structure.
 allowed-tools:
-- Read
-- Bash
-- Write
-- Edit
-- TodoWrite
+  - Read
+  - Bash
+  - Write
+  - Edit
+  - TodoWrite
 ---
 
 # Alfred EARS Authoring Guide
 
+## Skill Metadata
+| Field | Value |
+| ----- | ----- |
+| Allowed tools | Read (read_file), Write (write_file), Edit (edit_file), Bash (terminal), TodoWrite (todo_write) |
+| Auto-load | SessionStart (foundation bootstrap) |
+| Trigger cues | Requests to draft or refine EARS-style requirements, “write spec”, or “requirements format” cues. |
+
 ## What it does
 
 EARS (Easy Approach to Requirements Syntax) authoring guide for writing clear, testable requirements using 5 statement patterns.
 
 ## When to use
 
-- "SPEC 작성", "요구사항 정리", "EARS 구문"
+- Activates whenever the user asks to draft structured requirements or mentions EARS syntax.
+- “Writing SPEC”, “Requirements summary”, “EARS syntax”
 - Automatically invoked by `/alfred:1-plan`
 - When writing or refining SPEC documents
 
@@ -25,37 +34,65 @@ EARS (Easy Approach to Requirements Syntax) authoring guide for writing clear, t
 
 EARS provides 5 statement patterns for structured requirements:
 
-### 1. Ubiquitous (기본 요구사항)
-**Format**: 시스템은 [기능]을 제공해야 한다
-**Example**: 시스템은 사용자 인증 기능을 제공해야 한다
+### 1. Ubiquitous (Basic Requirements)
+**Format**: The system must provide [function]
+**Example**: The system must provide user authentication function
 
-### 2. Event-driven (이벤트 기반)
-**Format**: WHEN [조건]이면, 시스템은 [동작]해야 한다
-**Example**: WHEN 사용자가 로그인하면, 시스템은 JWT 토큰을 발급해야 한다
+### 2. Event-driven (event-based)
+**Format**: WHEN If [condition], the system must [operate]
+**Example**: WHEN When the user logs in, the system must issue a JWT token
 
-### 3. State-driven (상태 기반)
-**Format**: WHILE [상태]일 때, 시스템은 [동작]해야 한다
-**Example**: WHILE 사용자가 인증된 상태일 때, 시스템은 보호된 리소스 접근을 허용해야 한다
+### 3. State-driven
+**Format**: WHILE When in [state], the system must [operate]
+**Example**: WHILE When the user is authenticated, the system must allow access to protected resources
 
-### 4. Optional (선택적 기능)
-**Format**: WHERE [조건]이면, 시스템은 [동작]할 수 있다
-**Example**: WHERE 리프레시 토큰이 제공되면, 시스템은 새로운 액세스 토큰을 발급할 수 있다
+### 4. Optional (Optional function)
+**Format**: If WHERE [condition], the system can [operate]
+**Example**: If WHERE refresh token is provided, the system can issue a new access token
 
-### 5. Constraints (제약사항)
-**Format**: IF [조건]이면, 시스템은 [제약]해야 한다
-**Example**: IF 잘못된 토큰이 제공되면, 시스템은 접근을 거부해야 한다
+### 5. Constraints
+**Format**: IF [condition], the system SHOULD [constrain]
+**Example**: IF an invalid token is provided, the system SHOULD deny access
 
 ## Writing Tips
 
 ✅ Be specific and measurable
-✅ Avoid vague terms ("적절한", "충분한", "빠른")
+✅ Avoid vague terms (“adequate”, “sufficient”, “fast”)
 ✅ One requirement per statement
 ✅ Make it testable
 
 ## Examples
+```markdown
+- Scan standard documents and report missing sections.
+- Changed regulations are reflected in CLAUDE.md.
+```
+
+## Inputs
+- Project standard documents (e.g. `CLAUDE.md`, `.moai/config.json`).
+- Latest printouts from relevant sub-agents.
+
+## Outputs
+- Templates or policy summaries conforming to the MoAI-ADK standard.
+- Reusable rules/checklists.
+
+## Failure Modes
+- When required standard files are missing or have limited access rights.
+- When conflicting policies are detected and coordination is required.
+
+## Dependencies
+- There is great synergy when called together with cc-manager.
+
+## References
+- Mavin, A., et al. "Easy Approach to Requirements Syntax (EARS)." IEEE RE, 2009.
+- INCOSE. "Guide for Writing Requirements." INCOSE-TP-2010-006-02 (accessed 2025-03-29).
+
+## Changelog
+- 2025-03-29: Foundation skill templates have been enhanced to align with best practice structures.
 
-User: "JWT 인증 SPEC 작성해줘"
-Claude: (applies EARS patterns to structure authentication requirements)
 ## Works well with
 
 - moai-foundation-specs
+
+## Best Practices
+- When changing standards, the reason for change and supporting documents are recorded.
+- Follow the single source principle and avoid modifying the same item in multiple places.