메뉴
Code review Skill under PDD framework, verifying whether feature implementation meets development specifications and acceptance criteria. Automatically triggered after feature implementation or manually invoked. 支持中文触发：代码审查、代码评审、审查代码、检查代码、PDD代码审查。
Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.
SOC 직업 분류 기준
| name | pdd-code-reviewer |
| description | Code review Skill under PDD framework, verifying whether feature implementation meets development specifications and acceptance criteria. Automatically triggered after feature implementation or manually invoked. 支持中文触发：代码审查、代码评审、审查代码、检查代码、PDD代码审查。 |
| license | MIT |
| compatibility | Requires code implementation to be completed first |
| metadata | {"author":"neuqik@hotmail.com","version":"2.0","parent":"pdd-main"} |
Code Review - Verify whether feature implementation meets development specifications
Input:
Output:
| Version | Date | Changes |
|---|---|---|
| 2.1 | 2026-03-22 | Added collaboration with expert-code-quality: review first then analyze, quality issues do not block |
| 2.0 | 2026-03-21 | Added architect review invocation, integrated software-architect / system-architect |
| 1.0 | Early | Initial version |
When architecture issues are discovered during code review, invoke architect skills as needed:
| Architect Skill | Trigger Condition | Service Content |
|---|---|---|
| software-architect | Architecture deviation, interface design issues discovered | Architecture recommendations, design patterns |
| system-architect | System architecture issues discovered | System architecture review |
When technical issues are discovered during code review, invoke expert-xxx as needed:
| Expert Skill | Trigger Condition | Service Content |
|---|---|---|
| expert-code-quality | Code smells, refactoring needs discovered | Refactoring solutions, design patterns |
| expert-ruoyi | RuoYi framework usage issues discovered | Framework best practices |
| expert-mysql | SQL issues discovered | Optimization recommendations |
Must perform basic review:
Invoke architects as needed:
Invoke experts as needed:
Collect code files to be reviewed:
Read specification definitions from dev-specs/FP-{number}/spec.md.
Review against specifications item by item:
a. Interface Review
b. Business Logic Review
c. Data Model Review
Invoke expert-code-quality (if code quality issues discovered):
Invoke software-architect (if architecture issues discovered):
Output review report to docs/reviews/review-{timestamp}.md:
# Code Review Report
## Basic Information
| Item | Content |
|------|------|
| Feature Point | FP-XXX-NNN |
| Review Date | {date} |
| Reviewer | AI |
| Code Files | [...files...] |
## Review Results
### Passed Items
| Review Item | Status | Description |
|--------|------|------|
| Interface Design | ✓ | Conforms to specifications |
| ... | ... | ... |
### CRITICAL Issues
| No. | Issue Description | File | Recommendation |
|------|---------|------|------|
| 1 | ... | file.java | Fix solution |
### WARNING Issues
| No. | Issue Description | File | Recommendation |
|------|---------|------|------|
| 1 | ... | file.java | Optimization suggestion |
### SUGGESTION Issues
| No. | Issue Description | File | Recommendation |
|------|---------|------|------|
| 1 | ... | file.java | Refactoring suggestion |
## Architecture Review (if any)
### software-architect Review Comments
### system-architect Review Comments
## Conclusion
- [ ] Passed review
- [ ] Requires fix and re-review
Return issue list for fixing:
{
"critical": [...],
"warning": [...],
"suggestion": [...]
}
pdd-code-reviewer discovers architecture issues
│
├─► Determine architecture issue type
│ │
│ ├── Software architecture issues
│ │ │
│ │ ▼
│ │ software-architect consultation
│ │ │
│ │ ▼
│ │ Return architecture recommendations
│ │
│ └── System architecture issues
│ │
│ ▼
│ system-architect consultation
│ │
│ ▼
│ Return system recommendations
│
└─► Integrate into review report
Trigger condition: Module boundaries unclear discovered
Invoke: software-architect
Input:
- Problem description: Module A depends on Module C, violating layering principles
- Related code: [code snippet]
- Existing architecture: [module relationship diagram]
Return:
- Architecture issue analysis
- Refactoring recommendations
- Design pattern recommendations
Trigger condition: System scalability issues discovered
Invoke: system-architect
Input:
- Problem description: Current architecture cannot support future multi-tenant expansion
- Existing architecture: [architecture diagram]
- Expansion requirements: Support 100+ tenants
Return:
- Architecture evaluation
- Evolution roadmap
- Technical decision recommendations
| Skill | Positioning | Core Responsibility | Blocking |
|---|---|---|---|
| pdd-code-reviewer | Process compliance review | Verify whether code implements specification requirements | Critical issues block |
| expert-code-quality | Code quality deep analysis | Identify code smells, recommend design patterns, refactoring suggestions | Does not block process |
Code implementation completed
│
▼
┌─────────────────────────────────┐
│ pdd-code-reviewer │
│ (Compliance review) │
│ │
│ - Verify specification implementation │
│ - Check interface consistency │
│ - Verify business rules │
└─────────────────────────────────┘
│
├── Has Critical issues → Return to pdd-implement-feature for fixing
│
└── No Critical issues
│
▼
┌─────────────────────────────────┐
│ expert-code-quality │
│ (Quality deep analysis) │
│ │
│ - Code smell detection │
│ - Design pattern recommendations │
│ - Refactoring suggestions │
└─────────────────────────────────┘
│
▼
┌─────────────────────────────────┐
│ Generate quality improvement tasks │
│ (improvement-tasks.md) │
│ │
│ - Does not block current process │
│ - Record to future optimization list │
└─────────────────────────────────┘
│
▼
Enter pdd-verify-feature
| Issue Source | Level | Blocking | Handling Method |
|---|---|---|---|
| pdd-code-reviewer | Critical | ✅ Blocks | Must fix before continuing |
| pdd-code-reviewer | Warning | ❌ Does not block | Record, recommend fixing |
| pdd-code-reviewer | Suggestion | ❌ Does not block | Record, optional optimization |
| expert-code-quality | Any level | ❌ Does not block | Record to quality improvement list |
dev-specs/FP-{module}-{number}/improvement-tasks.md| Collaborative Skill | Collaboration Method | Input Data | Expected Output |
|---|---|---|---|
| software-architect | Consultation | Architecture issues | Architecture recommendations |
| system-architect | Consultation | System issues | System recommendations |
| expert-code-quality | Consultation | Code issues | Refactoring solutions |
| expert-ruoyi | Consultation | Framework issues | Best practices |
| pdd-implement-feature | Loop | Issue list | Fixed code |
| pdd-verify-feature | Sequential | Code passed review | Acceptance report |
Architecture constraint enforcement skill that monitors code for violations of preset invariants and architectural boundaries. Automatically triggered when users need architecture checks, dependency checks, or module boundary checks. Core responsibility: Ensure architectural coherence and prevent architectural drift. Trigger scenarios: - User requests "check architecture constraints", "check dependency direction", "check module boundaries" - Called by pdd-entropy-reduction coordinator - Automatically triggered after code commits 支持中文触发：架构约束检查、依赖检查、模块边界检查、架构强制、架构漂移、PDD架构约束。
Automated refactoring expert skill that transforms collected quality improvement tasks into concrete code operations. Automatically triggers when users need code refactoring, duplicate elimination, or complexity simplification. Core responsibility: Initiate targeted refactoring PRs regularly in a "small debt repayment" manner to prevent technical debt accumulation. Trigger scenarios: - User requests "refactor code", "eliminate duplicates", "simplify code" - Called by pdd-entropy-reduction coordinator - Refactoring suggestions passed from expert-entropy-auditor 支持中文触发：自动重构、代码重构、消除重复、简化代码、重构专家、PDD重构。
Entropy audit skill that discovers the gap between "design intent" and "code implementation". Automatically triggered when users need technical debt audit, PRD and code consistency check, or AI residue detection. Core responsibility: Identify "AI residue" scattered in the codebase and suggest consolidating them into shared utility packages. Trigger scenarios: - User requests "audit technical debt", "check PRD consistency", "detect AI residue" - pdd-entropy-reduction coordinator calls - Scheduled trigger (recommended weekly) 支持中文触发：熵增审计、技术债务审计、PRD一致性检查、AI残渣检测、PDD熵审计。
PDD Entropy Reduction Agent, continuously monitors and repays technical debt to prevent system decay. Automatically triggered when users need code cleanup, documentation updates, technical debt management, architecture alignment, entropy reduction, or garbage collection. Core Objective: Combat system "entropy increase" and "decay" by periodically running agents to discover documentation inconsistencies or architecture constraint violations. Trigger Scenarios: - User requests "entropy reduction", "clean up technical debt", "code cleanup", "garbage collection" - User requests checking documentation and code consistency - User requests architecture constraint checking - User requests technical debt audit - Scheduled trigger (recommended weekly) - Automatic trigger after code commit (optional) - Automatic trigger after PR merge (optional) 支持中文触发：熵减、清理技术债务、代码清理、文档更新、技术债务管理、架构对齐、垃圾回收、PDD熵减。
Activiti workflow engine expert, proficient in Activiti 7 Core/Cloud architecture, BPMN 2.0 specification, process design and development. Automatically triggered when users involve workflow design, process engine development, BPMN modeling, process deployment management and other issues. Provides BPMN modeling, best practices and solutions. 支持中文触发：工作流设计、流程引擎开发、BPMN建模、流程部署管理、Activiti、工作流、流程设计。
Code quality expert integrating Martin Fowler's refactoring techniques and GoF design patterns to help developers systematically improve code quality. Triggers when users ask about code review, refactoring, design patterns, code smells, SOLID principles, or software architecture improvements. 支持中文触发：代码审查、重构、设计模式、代码异味、SOLID原则、软件架构改进、代码质量。