一键导入
architecture-review
Use when reviewing architecture decisions. C4 model, dependency inversion, hexagonal boundaries.
用 Codex 或 Claude 帮你安装 复制这段 Prompt,粘贴到 Codex、Claude 或其他助手里,让它检查 Skill 页面并帮你完成安装。
菜单
Use when reviewing architecture decisions. C4 model, dependency inversion, hexagonal boundaries.
用 Codex 或 Claude 帮你安装 复制这段 Prompt,粘贴到 Codex、Claude 或其他助手里,让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
| name | architecture-review |
| description | Use when reviewing architecture decisions. C4 model, dependency inversion, hexagonal boundaries. |
| version | 1.0.0 |
| tags | ["architecture","review"] |
∀ plan reviews (standard, complex). ∀ PRs that touch module boundaries.
| Aspect | Status | Finding |
|---|---|---|
| Layer dependency | pass/fail | ... |
| Module boundaries | pass/fail | ... |
| Port coverage | pass/fail | ... |
| Cross-cutting concerns | pass/fail | ... |
Apply these structural tests during every architecture review:
Ask: "If I remove this module, does complexity scatter across the system or consolidate elsewhere?"
Identify modules with large surface area and low behavioral leverage. These are candidates for deepening — merging, hiding behind a narrower interface, or replacing with a library.
Locate seams where two modules cannot change independently. One adapter = hypothetical seam. Two adapters = real seam. Entangled units without a seam violate the boundary rule.
Flag modules that require heavy mocking or complex setup. "The interface is the test surface." Difficult testing often signals a shallow module or a missing port.
Document findings using domain terminology from CONTEXT.md or DOMAIN.md, not generic handler or service names.
Use during discuss-slice and plan-slice to stress-test a spec or plan against domain vocabulary before approval.
Use during research-slice to build disposable experiments that answer technical uncertainty.
Use during project:init to scaffold repo-level agent configuration (CLAUDE.md, AGENTS.md, tracker labels).
Use when debugging. 4-phase investigation, root cause, minimal fix.
Use when implementing features/fixes. Iron law: ¬∃ production code without failing test.
Use during discuss-slice to synthesize conversation context into a formal PRD with user stories, implementation decisions, testing decisions, and exclusions.