solid-review

Invocation

이 스킬은 두 가지 경로로 호출됩니다 — 어느 쪽이든 본 SKILL 본문의 절차를 그대로 실행합니다:

1. Claude Code 슬래시 — 사용자가 /solid-review [args...] 입력 (skill 의 user-invocable: true 가 슬래시 진입을 허용).
2. 타 에이전트 / Codex / Copilot CLI / Gemini CLI / SDK — Skill({ skill: "deep-work:solid-review", args: "..." }) 형태로 명시 invoke (cross-platform 표준 경로).

두 경로 모두 args 는 동일한 토큰 문자열로 전달되며, 본문 ($ARGUMENTS 자리) 의 파서가 동일하게 처리합니다.

Inputs (skill args)

인자	의미
(없음)	Auto-detect scope: 활성 세션의 changed files, 없으면 현재 디렉터리
<target>	File path / directory / glob pattern

빈 args / 매칭되지 않는 토큰 → 본문의 default 분기로 진입.

Prerequisites

이 entry skill 은 deep-work-orchestrator (Phase dispatch) 및 deep-work-workflow (reference skill — Phase 규약/Exit Gate/M3 envelope) 와 함께 동작합니다. 활성 deep-work 세션이 있을 때는 세션 state file (.claude/deep-work.<SESSION_ID>.md) 의 변수 (work_dir, current_phase, active_slice 등) 를 읽어 동작하며, 세션 외부에서도 standalone 실행이 가능한 경우 본문의 분기를 따릅니다.

Cross-platform self-containment: Claude Code 에서는 sibling skill 이 description 매칭으로 자동 로드됩니다. Codex / Copilot CLI / Gemini CLI / Agent SDK 에서 Skill() 로 호출 시 sibling auto-load 보장이 약할 수 있으므로, 본문은 self-contained 으로 보존되어 있습니다 — state file 해석, $ARGUMENTS 파싱, AskUserQuestion 분기, 출력 포맷이 인라인.

Quality Gate (v6.2.4) — /deep-test가 Advisory Gate로 자동 실행합니다. 특정 파일/디렉터리에 대한 독립 SOLID 검증이 필요할 때 직접 사용하세요. Standalone: /solid-review [target]

SOLID Design Review

You are performing a SOLID Design Review — analyzing code against the 5 SOLID design principles to evaluate design quality and suggest improvements.

Language

Detect the user's language from their messages or the Claude Code language setting. Output ALL user-facing messages in the detected language. The display templates below use Korean as the reference format — translate naturally to the user's language while preserving emoji, formatting, and structure.

Critical Constraints

• DO NOT modify any code files. This is a review-only operation.
• Read, analyze, and report findings.
• Save review results to file when in workflow mode.

Instructions

1. Determine operating mode

Resolve the current session's state file:

1. If DEEP_WORK_SESSION_ID env var is set → .claude/deep-work.${DEEP_WORK_SESSION_ID}.md
2. If .claude/deep-work-current-session pointer file exists → read session ID → .claude/deep-work.${SESSION_ID}.md
3. Legacy fallback → .claude/deep-work.local.md

Set $STATE_FILE to the resolved path.

Check if $STATE_FILE exists and has an active session (current_phase is not idle and not empty).

Workflow Mode (active deep-work session):

• Read work_dir from the state file
• Set WORK_DIR to the value of work_dir
• Read $WORK_DIR/plan.md to extract the list of files to review (from "Files to Modify" section)
• Read $WORK_DIR/research.md for architectural context (Executive Summary section only)
• Review scope: files listed in plan.md that were actually modified during implementation

Standalone Mode (no active session):

• If $ARGUMENTS is provided: use as target (file path, directory, or glob pattern)
• If $ARGUMENTS is empty: detect scope automatically:
  1. Check git diff --name-only HEAD~1 for recently changed files
  2. If not a git repo or no changes, use current directory
• Review scope: all code files in detected scope (exclude node_modules, .git, pycache, build, dist, etc.)

2. Collect review targets

Gather the list of files to review. For each file:

• Read the file contents
• Skip files that are clearly not code (README.md, .json config, .env, etc.)
• Skip files smaller than 5 lines (trivial)
• Skip auto-generated files (migrations, lock files, bundled output)

If the total number of files exceeds 20, prioritize:

1. Files with the most lines of code
2. Files with class/interface definitions
3. Files explicitly listed in plan.md (workflow mode)

Display progress:

SOLID 리뷰 대상: [N]개 파일
  - src/auth/service.ts (245 lines)
  - src/models/user.ts (180 lines)
  - ...

3. Analyze each principle

For each file (or logical group of related files), evaluate against all 5 SOLID principles.

Read ${CLAUDE_PLUGIN_ROOT}/skills/shared/references/solid-guide.md for the detailed checklist.

Analysis approach:

• Do NOT mechanically check every rule against every file
• Focus on violations that actually matter in the given context
• Consider the project's scale and maturity (KISS balance)
• A small utility script doesn't need DIP — flag it only in core domain logic

For each principle, assign one of:

• 준수: No violations found, or principle is not applicable
• 개선 권장: Minor violations that would improve maintainability
• 위반: Clear violations that will cause maintenance problems

4. Generate scorecard

Per-file scorecard (for each reviewed file):

### [filename] ([N] lines)

| 원칙 | 상태 | 요약 |
|------|------|------|
| SRP  | [status] | [1-line summary] |
| OCP  | [status] | [1-line summary] |
| LSP  | [status] | [N/A or finding] |
| ISP  | [status] | [1-line summary] |
| DIP  | [status] | [1-line summary] |

Overall scorecard:

## 종합 SOLID 스코어카드

| 원칙 | 전체 상태 | 위반 파일 수 | 핵심 발견 |
|------|----------|-------------|----------|
| SRP  | [status] | N/M         | [most common issue] |
| OCP  | [status] | N/M         | — |
| LSP  | [status] | N/M         | — |
| ISP  | [status] | N/M         | [most common issue] |
| DIP  | [status] | N/M         | [most common issue] |

**총점**: N/5 원칙 준수
**판정**: 개선 권장 (Advisory — 워크플로우 차단 없음)

5. Generate refactoring suggestions

For each violation or improvement finding, provide a concrete refactoring suggestion. Limit to top 5 suggestions sorted by impact:

## 리팩토링 제안

### 1. [SRP] PlayerController.cs — 책임 분리
**현재**: 이동, 입력, UI 업데이트가 한 클래스에 혼재
**제안**: PlayerMover, PlayerInput, PlayerUI로 분리
**우선순위**: 높음

### 2. [DIP] AuthService.ts — 추상화 도입
**현재**: `new DatabaseClient()` 직접 생성
**제안**: `IDatabaseClient` 인터페이스 추출, 생성자 주입
**우선순위**: 중간

6. AI 프롬프트 개선 제안 (워크플로우 모드, 선택적)

If running in workflow mode and clear SOLID violations were found, read ${CLAUDE_PLUGIN_ROOT}/skills/shared/references/solid-prompt-guide.md and suggest how plan.md could be improved:

## AI 프롬프트 개선 제안

다음 plan 작성 시 아래 조건을 추가하면 SOLID 위반을 사전에 방지할 수 있습니다:
- "각 클래스는 단일 책임만 담당하도록 분리할 것 (SRP)"
- "새 기능 추가 시 기존 코드 수정 없이 확장 가능한 구조로 설계할 것 (OCP)"
- "구체 클래스 대신 인터페이스에 의존하도록 구현할 것 (DIP)"

7. Save results

Workflow Mode:

• Write results to $WORK_DIR/solid-review.md
• Display summary in terminal

Standalone Mode:

• Display full results in terminal
• Ask user: "리뷰 결과를 파일로 저장할까요? (기본: 아니오)"
• If yes, save to ./solid-review.md

8. Workflow integration (workflow mode only)

If called as a Quality Gate during Test phase:

• Record results in quality-gates.md as Advisory entry
• Violations do NOT block the workflow — record warning only

If called outside the Test phase:

• Still run the review
• Note: "deep-work 워크플로우 활성 상태 — 결과가 $WORK_DIR/solid-review.md에 저장됩니다"