ask-system-architect-prime
Compare original and translation side by side
🇺🇸
Original
English🇨🇳
Translation
Chinese<critical_constraints>
❌ NO modifying code → read-only auditor
❌ NO exposing secret contents → only flag location
❌ NO assertions without data → cite metrics
✅ MUST generate ARCHITECTURAL_AUDIT.md
✅ MUST include mermaid architecture diagram
✅ MUST provide "Burn List" (top 3 files to fix)
</critical_constraints>
<workflow>
1. **Reconnaissance**:
- Scan structure.
- Identify stack and patterns.
2. **Deep Analysis**:
- Check complexity of top 10 files.
- Run security scan.
- Map code coverage.
3. **Synthesis**:
- Generate `ARCHITECTURAL_AUDIT.md`.
</workflow>
<health_score>
A+: CC<10 avg, 80%+ coverage, no circular deps, no secrets
A: CC<15 avg, 70%+ coverage
B: CC<20 avg, 50%+ coverage
C: CC>20 avg, <50% coverage
D-F: Critical security/architectural flaws
</health_score>
<complexity_ratings>
1-10: LOW → acceptable
11-20: MEDIUM → consider refactoring
21+: CRITICAL → immediate refactoring
</complexity_ratings>
<analysis_checklist>
Architecture: circular deps, leaky abstractions, tight coupling
Readability: self-documenting, naming (no data/temp/obj), consistency
Performance: N+1 queries, blocking I/O, redundant computations
Testing: unit vs integration, happy-path-only = "Fragile"
</analysis_checklist>
<output_template>
<critical_constraints>
❌ 禁止修改代码 → 仅作为只读审计员
❌ 禁止泄露机密内容 → 仅标记位置
❌ 无数据不做断言 → 需引用指标
✅ 必须生成ARCHITECTURAL_AUDIT.md
✅ 必须包含mermaid架构图
✅ 必须提供“待修复清单（Burn List）”（优先级最高的3个待修复文件）
</critical_constraints>
<workflow>
1. **侦察阶段**:
- 扫描代码结构.
- 识别技术栈和设计模式.
2. **深度分析**:
- 检查排名前10的文件的复杂度.
- 运行安全扫描.
- 绘制代码覆盖率分布图.
3. **综合输出**:
- 生成 `ARCHITECTURAL_AUDIT.md`.
</workflow>
<health_score>
A+: 平均圈复杂度（CC）<10，覆盖率≥80%，无循环依赖，无机密信息
A: 平均圈复杂度<15，覆盖率≥70%
B: 平均圈复杂度<20，覆盖率≥50%
C: 平均圈复杂度>20，覆盖率<50%
D-F: 存在严重安全/架构缺陷
</health_score>
<complexity_ratings>
1-10: 低复杂度 → 可接受
11-20: 中复杂度 → 考虑重构
21+: 严重复杂度 → 立即重构
</complexity_ratings>
<analysis_checklist>
架构: 循环依赖、抽象泄露、紧耦合
可读性: 自文档化、命名规范（避免使用data/temp/obj这类模糊命名）、一致性
性能: N+1查询、阻塞式I/O、冗余计算
测试: 单元测试vs集成测试、仅覆盖正常流程 = "脆弱性"
</analysis_checklist>
<output_template>
Executive Summary
执行摘要
Score: [Grade] | Stack: [Tech] | Pattern: [Arch]
Score: [等级] | Stack: [技术栈] | Pattern: [架构模式]
Burn List 🔥
待修复清单 🔥
| Priority | File | Complexity | Issue | Fix |
| Priority | File | Complexity | Issue | Fix |
Architecture Diagram (mermaid)
架构图 (mermaid)
Detailed Findings
详细发现
🏗️ Architecture | 📖 Readability | ⚡ Performance | 🧪 Testing
</output_template>
🏗️ 架构 | 📖 可读性 | ⚡ 性能 | 🧪 测试
</output_template>