junnv93
25 verified skills
manage-skills
Analyzes session changes to detect missing verification skills. Dynamically discovers existing skills, creates new skills or updates existing ones, and manages CLAUDE.md skill references. Use when adding new patterns/modules that may need verification coverage, or when maintaining skill consistency. 스킬 관리, 검증 스킬 누락 탐지, CLAUDE.md 업데이트.
data-ai
generate-prompts
Scan codebase for actionable issues and generate verified harness prompts. Auto-archive completed items and filter false positives. Use when: user asks to scan codebase, generate new prompts, review existing prompts, clean up completed items, or says "프롬프트 생성", "코드베이스 스캔", "새 프롬프트", "/generate-prompts". Also triggers on "프롬프트 리뷰", "프롬프트 정리".
development
nextjs-16
Next.js 16 App Router reference guide (NOT a verification skill — use /verify-nextjs for validation). Use when creating or modifying pages, layouts, routes, Server Actions, or Server Components. Covers params Promise pattern, PageProps typing, useActionState (replaces useFormState), Server/Client component boundaries, and Cache Components. 페이지/레이아웃/라우트/Server Action 작성 시 참조.
tools
playwright-e2e
Playwright E2E test workflow orchestrator — covers planning, code generation, execution, healing, and verification. Coordinates playwright-test MCP server's planner/generator/healer agents with sequential execution strategy to prevent browser resource contention. Automatically applies project-specific patterns (auth fixtures, storageState, role-based testing). Use whenever the user requests E2E tests, Playwright tests, test plans, test generation, page tests, feature tests, role-based tests, or permission tests. Trigger on: "E2E 테스트", "Playwright 테스트", "테스트 계획", "테스트 생성", "기능 테스트", "권한 테스트".
tools
verify-cas
Verifies CAS (Optimistic Locking) pattern compliance — version field in DTOs, VersionedBaseService usage, cache invalidation on 409 conflict, frontend VERSION_CONFLICT error handling. Run after adding/modifying state-change endpoints.
development
verify-frontend-state
Verifies frontend state management patterns + performance anti-patterns — TanStack Query for server state (no useState), onSuccess setQueryData prohibition, dynamic imports for code splitting, Server/Client component separation. Run after adding/modifying components or hooks.
tools
verify-hardcoding
SSOT 소스가 존재하는 값의 하드코딩을 탐지합니다. API 경로, queryKeys, 환경변수, 캐시 키, 토큰 TTL, 페이지 옵션 등. 기능 구현 후 사용.
development
verify-security
보안 설정을 검증합니다. Helmet CSP, Next.js Security Headers, PermissionsGuard 모드, @Public 남용 검사.
testing
harness
3-Agent harness orchestrator (Planner → Generator → Evaluator loop). Auto-select execution mode based on task complexity. Reuse existing verify-*/review-* skills as Evaluator infrastructure. Trigger on "하네스", "/harness", "harness mode", or when starting non-trivial multi-file implementation tasks.
testing
verify-auth
Verifies server-side authentication/authorization pattern compliance — req.user.userId extraction (no body userId), @RequirePermissions decorator, @AuditLog decorator, JwtAuthGuard coverage. Run after adding/modifying controller endpoints.
testing
verify-nextjs
Verifies Next.js 16 pattern compliance — await params/searchParams, PageProps typing, useActionState, Server/Client component boundaries, dynamic imports. Run after adding/modifying pages or layouts.
tools
verify-ssot
SSOT(Single Source of Truth) 임포트 소스를 검증합니다. 타입/enum/상수가 올바른 패키지에서 임포트되는지 확인. 타입/enum 추가/수정 후 사용.
testing
verify-zod
Verifies Zod validation pattern compliance — ZodValidationPipe usage (no class-validator), versionedSchema inclusion in state-change DTOs, controller pipe application, query DTO consistency. Run after adding/modifying DTOs or controller endpoints.
testing
verify-workflows
Verifies cross-feature workflow E2E test coverage against critical-workflows.md checklist. Checks WF-01~WF-35 coverage, step completeness, role correctness, side-effect verification, and status transition assertions. Run after adding workflow tests or before PR.
testing
equipment-management
Equipment Management System development guide based on UL-QP-18 laboratory equipment management procedures. Use this skill whenever implementing features related to: equipment CRUD, calibration records/plans/approval workflows, checkout/rental/import flows, non-conformance management, disposal requests, inspection (중간점검/자체점검), calibration factors, software registry (UL-QP-18-07), management number system, role-based access control (4 roles), or approval processes (1-step/2-step/3-step). Also use when working on equipment history cards (시험설비이력카드 UL-QP-18-02), frontend UI with Next.js 16 App Router, or NextAuth token management. Trigger on keywords: 장비, equipment, calibration, 교정, checkout, 반출, non-conformance, 부적합, disposal, 폐기, approval, 승인.
development
.claude/skills/git-commit
--- name: git-commit description: Analyzes git changes to generate conventional commit messages and auto-selects workflow (direct push to main vs branch+PR) based on change size. Use when the user wants to commit, push, save changes, or prepare a PR. Trigger on: "커밋해줘", "커밋하고 푸시해줘", "변경사항 정리해줘", "git commit", "코드 올려줘", "commit", "push", "save changes", "create PR", "세션 마무리해줘". --- # Git 커밋 + 배포 워크플로우 ## 개요 변경사항을 분석하여 커밋 메시지를 생성하고, **변경 규모에 따라 워크플로우를 자동 결정**합니다: - **작은 수정** → main에 직접 커밋+푸시 -
data-ai
.claude/skills/review-architecture
--- name: review-architecture description: Architecture-level code review — traces changes across DB→Service→Controller→DTO→Hook→Component→Cache layers, verifying cross-module pattern consistency, CAS coherence, cache invalidation correctness, security layers, and performance anti-patterns. Unlike verify-* skills (rule-based automated checks), this provides senior architect-level design judgment. Use after adding new modules, implementing complex state-change logic, multi-step approval flows, or
development
review-design
Frontend design quality review + improvement wireframe HTML generation. Scores existing components/pages against 10 anti-patterns (card soup, spacing rhythm, typography drama, depth/elevation, state indicators, motion, tables, accessibility, empty states, design token usage) and generates reference wireframe HTML for improvement direction. Use when the user asks for design review, UI review, UI quality assessment, design scoring, or mentions: "디자인 리뷰", "UI 리뷰", "촌스러운", "세련되게", "다듬어줘", "깔끔하게", "프리미엄", "카드 수프", "visual hierarchy", "wireframe". This skill evaluates existing UI quality — new page implementation is handled separately.
development
verify-design-tokens
Verifies Design Token 3-Layer architecture compliance — no transition-all (use specific properties), focus-visible over focus, correct import paths, layer reference chain (primitive→semantic→component). Run after adding/modifying components.
testing
verify-e2e
Verifies E2E test code compliance AND architectural coverage — auth fixture usage, locator patterns, SSOT constants, test isolation, plus architecture-level checks (CAS VERSION_CONFLICT scenarios, cache invalidation after mutation, site access control on mutations). Run after adding/modifying E2E test specs.
development
verify-filters
Verifies URL-driven filter SSOT pattern compliance — filter-utils required exports, hook existence, page.tsx server-side parsing, no useState for filter state (URL params are SSOT). Run after adding/modifying list page filters.
testing
verify-i18n
Verifies i18n translation consistency — en/ko key pair matching, no empty translations, namespace references, Zod schema hardcoded message detection, dynamic i18n key coverage, routeMap↔navigation.json sync. Run after adding/modifying messages/{en,ko}/*.json or route-metadata.ts.
development
.claude/skills/verify-implementation
--- name: verify-implementation description: Runs all verify-* skills sequentially to produce a unified verification report. Use instead of running individual verify skills one by one. Run before PR creation, after feature implementation, or during code review. Trigger on: "전체 검증", "통합 검증", "PR 준비", "full verification", "run all checks", "verify everything". disable-model-invocation: true argument-hint: '[선택사항: 특정 verify 스킬 이름]' --- # 구현 검증 ## 목적 프로젝트에 등록된 모든 `verify-*` 스킬을 순차적으로 실행하여 통합 검증을
development
verify-seed-integrity
테스트 시드 인프라의 3자 SSOT 정합성을 검증합니다 — seed-data 파일 ↔ seed-test-new.ts 의 truncate/insert wiring ↔ verification.ts 의 count check 가 한 세트로 움직여야 함. 새 seed 파일 추가 후, verification.ts 편집 후 사용.
testing
verify-sql-safety
Verifies SQL safety — LIKE wildcard escaping, N+1 query pattern detection, COUNT(DISTINCT) for fan-out JOINs, RBAC INNER JOIN enforcement. Run after adding/modifying search or list API endpoints.
development