By chkim-su
MCP-based symbolic refactoring plugin with context isolation, duplicate detection, and intelligent feature injection.
Serena MCP-based SOLID analysis. Symbol-level analysis for accurate violation detection and auto-fix paths.
Quick code quality audit. Verifies code quality before and after refactoring.
Systematic debugging workflow with parallel agent exploration, root cause analysis, fix strategy design, and verification. Adapted from feature-dev methodology.
Duplicate code and role detection. Identifies clone code, similar functions, and duplicate constants, suggesting consolidation approaches.
Execute real user behavior simulation tests. UI interactions (clicks, inputs, drag-drop), API sequences, and scenario-based E2E testing using Playwright MCP.
Executes code injection based on implementation plan. Safely inserts new code, handles imports, and verifies successful injection using Serena MCP.
Generates creative and unconventional implementation approaches including API bypasses, technology combinations, and constraint-ignoring experiments. Helper agent for Phase 3 of the laboratory workflow.
Deeply analyzes codebase to trace bug execution paths, map data flows, and identify affected components. Uses Serena MCP for symbol-level investigation.
Designs fix strategies for identified bugs with risk assessment, side effect analysis, and implementation blueprints. Provides multiple approaches from minimal to comprehensive.
Verifies bug fixes by checking direct resolution, regression testing, and edge case coverage. Uses confidence-based reporting to filter false positives.
Systematic debugging workflow with parallel agent exploration, root cause analysis, and fix verification. Adapted from feature-dev methodology for bug investigation.
Code duplication detection rules. Patterns and thresholds for identifying clone code, role duplication, and variable duplication. Guide for removing duplicates through refactoring.
Rules and patterns for safely injecting new features into existing codebases. Ensures consistency with project conventions and SOLID principles.
Scientific experimentation patterns for exploring implementation approaches through hypothesis testing, parallel PoC execution, and rigorous verification. Used by laboratory command for systematic research workflows.
Project knowledge extraction and graph structure rules. Defines how to analyze, document, and maintain project architecture knowledge using Serena MCP.
Uses power tools
Uses Bash, Write, or Edit tools
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimBased on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
Serena MCP 기반 심볼릭 리팩토링 플러그인입니다.
v2.4.0 - 데이터 드리븐 아키텍처, 직접 MCP 통합
Main Session
├─ MCP 도구로 코드 데이터 수집 (find_symbol, get_symbols_overview 등)
└─ 수집된 데이터를 에이전트에게 전달
↓
Specialized Agents (도구 없음, 분석 전담)
├─ serena-solid-analyzer: SOLID 위반 분석
├─ refactor-planner: 리팩토링 계획 수립
├─ duplicate-detector: 중복 코드 탐지
└─ feature-planner: 기능 구현 계획
장점:
Serena daemon과 직접 통합:
activate_project)| 명령어 | 설명 | 워크플로우 단계 |
|---|---|---|
/serena-refactor:analyze | SOLID 분석 실행 | 1단계: 분석 |
/serena-refactor:plan | 리팩토링 계획 수립 | 2단계: 계획 |
/serena-refactor:refactor | 전체 리팩토링 워크플로우 | 3단계: 실행 |
/serena-refactor:audit | 코드 품질 감사 | 4단계: 검증 |
/serena-refactor:rename | 안전한 심볼 이름 변경 | 단독 작업 |
/serena-refactor:extract | 메서드/클래스 추출 | 단독 작업 |
/serena-refactor:detect-duplicates | 중복 코드 탐지 | 분석 단계 |
/serena-refactor:inject | 지능형 기능 주입 | 기능 추가 워크플로우 |
/serena-refactor:analyze → .refactor-analysis-done
↓
/serena-refactor:plan → .refactor-plan-approved
↓ (Hook: 선행 조건 검증)
/serena-refactor:refactor → .refactor-execution-done
↓ (필수)
/serena-refactor:audit → .refactor-audit-passed (PASS 시)
↓
git commit
knowledge-extractor → .inject-knowledge-extracted
↓
feature-planner → .inject-plan-approved
↓ (Hook: 계획 승인 확인)
code-injector → .inject-execution-done
↓
검증 및 git commit
차단형 Hook (선행 조건 미충족 시 실행 거부):
serena-refactor-executor: 분석 + 계획 완료 확인code-injector: 기능 계획 승인 확인경고형 Hook (정보 제공):
refactor-planner: 분석 완료 권장feature-planner: 지식 추출 완료 권장| 파일 | 생성 시점 | 의미 |
|---|---|---|
.refactor-analysis-done | analyze/detect-duplicates 완료 | 분석 완료 |
.refactor-plan-approved | plan 완료 | 계획 승인 (자동) |
.refactor-execution-done | refactor 완료 | 실행 완료 |
.refactor-audit-passed | audit PASS | 품질 검증 통과 |
.inject-knowledge-extracted | knowledge-extractor 완료 | 지식 추출 완료 |
.inject-plan-approved | feature-planner 완료 | 기능 계획 승인 |
.inject-execution-done | code-injector 완료 | 코드 주입 완료 |
serena-solid-analyzer: SOLID 원칙 위반 분석 (데이터 전용)duplicate-detector: 코드 클론 탐지 (데이터 전용)refactor-planner: 리팩토링 계획 수립 (데이터 전용)feature-planner: 기능 구현 계획 (데이터 전용)serena-refactor-executor: 리팩토링 실행 (계획 기반)code-injector: 코드 주입 (계획 기반)refactor-auditor: 품질 검증 (실행 결과 분석)knowledge-extractor: 프로젝트 지식 그래프 생성debug-explorer: 버그 탐색 및 실행 경로 추적debug-strategist: 수정 전략 설계 (최소/포괄/방어적)debug-verifier: 수정 결과 검증 (직접/회귀/엣지케이스)e2e-test-runner: Playwright 기반 사용자 시뮬레이션 테스트| 스킬 | 용도 |
|---|---|
solid-design-rules | SOLID 원칙 및 TDD 규칙 |
serena-refactoring-patterns | Serena 도구 기반 리팩토링 패턴 |
duplicate-detection-rules | 코드 클론 탐지 및 수정 패턴 |
project-knowledge-graph | 프로젝트별 누적 지식 |
feature-injection-rules | 코드 주입 템플릿 및 가이드 |
find_symbol: 심볼 검색 및 본문 읽기find_referencing_symbols: 참조 추적get_symbols_overview: 파일 구조 분석search_for_pattern: 패턴 검색replace_symbol_body: 심볼 본문 교체rename_symbol: 안전한 이름 변경 (모든 참조 자동 업데이트)insert_before/after_symbol: 심볼 삽입replace_content: 정규식 기반 콘텐츠 교체activate_project: 프로젝트 전환list_memories: 메모리 목록read_memory / write_memory: 프로젝트 지식 저장플러그인 설치 시 SessionStart hook이 Serena MCP 자동 설정:
# 플러그인 설치
claude plugin add local refactor
# Claude Code 재시작 → Serena MCP 자동 등록
# Serena MCP 등록
claude mcp add --transport stdio --scope user serena -- \
uvx --from git+https://github.com/oraios/serena serena start-mcp-server
# 플러그인 설치
cd ~/.claude/plugins/local/
git clone <this-repo> serena-refactor
Multi-command workflow system for creating Claude Code plugin components with daemon-owned policy and explicit phase transitions
Strict migration plugin with bootstrap requirement: /bootstrap must run first, manual workarounds blocked, enforcement scripts must exist for /migrate to work
Oracle-powered autonomous development pipelines with Gemini-Claude orchestration
npx claudepluginhub chkim-su/serena-refactorUltra-compressed communication mode. Cuts 65% of output tokens (measured) while keeping full technical accuracy by speaking like a caveman.
Multi-model consensus engine integrating OpenAI Codex CLI, Gemini CLI, and Claude CLI for collaborative code review and problem-solving.
Comprehensive UI/UX design plugin for mobile (iOS, Android, React Native) and web applications with design systems, accessibility, and modern patterns
Unified capability management center for Skills, Agents, and Commands.
Memory compression system for Claude Code - persist context across sessions
Standalone image generation plugin using Nano Banana MCP server. Generates and edits images, icons, diagrams, patterns, and visual assets via Gemini image models. No Gemini CLI dependency required.