[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M6E1SX5DP80CKTMNQET1M3S1":3,"skill-same:01M6E1SX5DP80CKTMNQET1M3S1":105},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":16,"categories":17,"keywords":23,"publishedAt":44,"linkOnly":45,"markdownContent":46,"markdownBody":47,"frontMatter":48,"refName":52,"commitSha":53,"skillDirectory":54,"forksCount":55,"repositoryHtmlUrl":56,"repositoryHomepageUrl":57,"ownerHtmlUrl":58,"ownerBlogUrl":59,"ownerDescription":6,"updatedAt":60,"install":61,"resources":74},"01M6E1SX5DP80CKTMNQET1M3S1","plugin-test",null,"Run NCC plugin E2E tests. Supports quick (4 projects), full (4+monorepo+modifications), install-test (npx vs marketplace with non-spec-dev focus), or custom test modes. Validates spec→dev flow, skills, hooks, code quality, security, a11y, and token efficiency.","ByeongminLee\u002Fnextjs-claude-code","ByeongminLee","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F71276006?v=4","01M6E1S467E7GC0NE144SWHC0V","Byeongmin Lee","01M6E1S4AV3W0B08Z2YBK26VG9",3,"MIT","https:\u002F\u002Fgithub.com\u002FByeongminLee\u002Fnextjs-claude-code\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Fplugin-test\u002FSKILL.md",[18],{"categoryNo":19,"categoryName":20,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000008","测试与质量","SCENARIO","使用场景",[24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43],"test","plugin","plugin test","phase2","security","phase1","token","project","efficiency","focus","hooks","install","orchestrator","monorepo","projects","flow","plugin e2e","code","run","e2e tests","2026-11-10 10:12:30",false,"---\nname: plugin-test\ndescription: Run NCC plugin E2E tests. Supports quick (4 projects), full (4+monorepo+modifications), install-test (npx vs marketplace with non-spec-dev focus), or custom test modes. Validates spec→dev flow, skills, hooks, code quality, security, a11y, and token efficiency.\nargument-hint: \"--quick | --full | --install-test | --phase1 | --phase2 | --security | --a11y | --codequality | [project-name,...] | (no args = suggest mode)\"\ncontext: fork\nagent: plugin-test-orchestrator\n---\n\n# \u002Fplugin-test — NCC Plugin E2E Test Suite\n\n## Usage\n\n```\n\u002Fplugin-test                    # Interactive — suggest a test mode based on recent changes\n\u002Fplugin-test --quick            # Phase 1 only: 4 diverse projects × 4 features each\n\u002Fplugin-test --full             # Phase 1 + Phase 2: full suite including monorepo + cross-feature mods\n\u002Fplugin-test --install-test     # Install test: 2 projects (npx + marketplace), 5 features each, non-spec-dev focus\n\u002Fplugin-test --phase1           # Phase 1 only (same as --quick)\n\u002Fplugin-test --phase2           # Phase 2 only: 10-feature monorepo (solo + team)\n\u002Fplugin-test --security         # Security-focused: run security-reviewer on all generated code\n\u002Fplugin-test --a11y             # Accessibility-focused: run browser-tester a11y checks on UI projects\n\u002Fplugin-test --codequality      # Code quality audit: deep analysis of all generated source files\n\u002Fplugin-test --token             # Token optimization audit: run token-optimizer on plugin source\n\u002Fplugin-test healthcare,marketplace   # Run specific projects only\n\u002Fplugin-test --create              # \u002Fcreate pipeline test: new + existing project ideation\n```\n\n## Test Modes\n\n### Phase 1: Quick Validation (--quick, --phase1)\n\n4 diverse projects, each with 4 features + extra agent verification.\n\n| Project | Stack | Extra Agents |\n|---------|-------|-------------|\n| `healthcare` | Prisma + next-auth + zod + vitest | `\u002Fsecurity`, `\u002Freview` |\n| `design-system` | framer-motion + playwright + storybook | `browser-tester`, `\u002Fqa --a11y` |\n| `realtime-collab` | drizzle + tanstack-query + zustand + msw | `performance-optimizer`, `\u002Floop` |\n| `marketplace` | Prisma + stripe + ai-sdk + @vercel\u002Fblob | `\u002Fsecurity --diff`, `\u002Freview`, `\u002Fstatus` |\n\nPer-project: NCC install → \u002Finit → 4× spec→dev (3 tasks each) → extra agent runs → TEST_LOG.md\n\nReviewer scores 15 categories:\n1-10: Installation, Hooks, Skills, Spec, Design, PLAN, TDD, Token, Code, Flow\n11: Rules size (\u003C200 lines each)\n12: Skill selection accuracy (manifest vs package.json)\n13: Extra agent execution results\n14: Document sync (CONTEXT.md\u002Fdesign.md consistency)\n15: Code quality detail (error handling, Zod, composition, Server\u002FClient)\n\n### Install Test (--install-test)\n\nTests the **published version** of NCC using two real installation methods. Primary focus: **non-spec-dev features**.\n\n| Project | Install Method | Stack | Features | Non-Spec-Dev Focus |\n|---------|---------------|-------|----------|-------------------|\n| `npx-healthcare` | `npx nextjs-claude-code@latest` | Prisma + next-auth + zod + vitest | 5 features | \u002Finit, \u002Fsecurity, \u002Flog, \u002Fcommit, \u002Frule, \u002Ftest, hooks, skills |\n| `mkt-marketplace` | `\u002Fplugin marketplace add` + `\u002Fplugin install` | Prisma + stripe + ai-sdk + @vercel\u002Fblob | 5 features | \u002Finit, \u002Freview, \u002Fstatus, \u002Fdebug, \u002Fqa, \u002Floop, \u002Fcommit, hooks, skills |\n\nPer-project: Install → \u002Finit → 5× spec→dev (2-3 tasks each) → non-spec-dev tests → TEST_LOG.md\n\nReviewer scores 20 categories (base 15 + 5 install-test categories):\n16: Init Completeness (PROJECT.md, ARCHITECTURE.md, STATE.md quality)\n17: Security & Log Setup (SECURITY_STRATEGY.md, LOG_STRATEGY.md, audit results)\n18: Git Workflow (\u002Fcommit conventional format + REQ linking)\n19: Debug & Loop (DEBUG.md hypothesis-driven, LOOP_NOTES.md tracking, \u002Fstatus read-only)\n20: Installation Method Fidelity (npx: local files + manifest; marketplace: namespace + CLAUDE_PLUGIN_ROOT)\n\n### Phase 2: Large-scale Monorepo (--phase2)\n\n2 identical CRM projects (solo vs team mode), each with:\n- 10 features in dependency chain\n- 5 cross-feature modifications (api-layer error format, RBAC expansion, tagging, cursor pagination, custom stages)\n- Document Sync verification (CONTEXT.md, design.md, history entries)\n- Solo vs Team comparison\n\nFeatures: api-layer → auth-system → contact-management → deal-pipeline → email-integration → activity-feed → reporting → team-workspace → notification-center → admin-settings\n\n### \u002Fcreate Pipeline Test (--create)\n\nTests the `\u002Fcreate` ideation pipeline on both new and existing projects.\n\n| Project | Scenario | Validation |\n|---------|----------|-----------|\n| `create-new` | Empty project → \u002Finit → \u002Fcreate \"AI recipe app\" | VISION.md, C-REVIEW.md, DECISION.md exist in spec\u002Fcreate\u002F |\n| `create-existing` | Healthcare project → \u002Fcreate \"notification system\" | Reads spec\u002FPROJECT.md context, outputs to spec\u002Fcreate\u002F |\n\nPer-project: NCC install → \u002Finit → \u002Fcreate → verify output docs → token isolation check → spec conversion test\n\nReviewer checks 5 categories:\n1. Document Quality: VISION.md has Problem\u002FDemand\u002FTarget\u002FValue sections\n2. C-Level Coverage: C-REVIEW.md has all 5 assessments (CEO\u002FCTO\u002FCPO\u002FCMO\u002FCDO) with verdicts\n3. Decision Clarity: DECISION.md has chosen approach + key decisions\n4. Token Isolation: After \u002Fcreate, run \u002Fspec and verify spec-writer does NOT read spec\u002Fcreate\u002F\n5. Spec Conversion: If converted, output is valid \u002Fspec command format\n\n### Full Suite (--full)\n\nRuns Phase 1, analyzes issues, then runs Phase 2. Generates consolidated REPORT with cross-phase comparison.\n\n### Security Audit (--security)\n\nFor each generated project:\n- Read installed `security-reviewer` agent\n- Scan all API routes for OWASP Top 10: injection, auth bypass, sensitive data exposure, CSRF\n- Check MSW handlers for auth mock bypass patterns\n- Check Zod validation coverage\n- Output: `SECURITY_REVIEW.md` per project\n\n### Accessibility Audit (--a11y)\n\nFor UI-heavy projects (design-system or any with UI components):\n- Read installed `browser-tester` agent\n- WCAG 2.1 AA checklist per component\n- Check: ARIA attributes, keyboard nav, focus management, color contrast\n- Check: prefers-reduced-motion respect, semantic HTML, role attributes\n- Output: `ACCESSIBILITY_REVIEW.md` per project\n\n### Code Quality Audit (--codequality)\n\nDeep source code analysis across all projects:\n- Error handling grade (try\u002Fcatch, error classification, no silent swallowing)\n- Zod validation grade (boundary validation, schema completeness)\n- TypeScript strictness (no any, no as unknown, proper inference)\n- Composition patterns (Context vs prop drilling, compound components)\n- Server\u002FClient discipline ('use client' minimization)\n- DRY compliance (no duplicate logic across routes)\n- Named constants (no magic numbers\u002Fstrings)\n- Output: code quality table in REPORT\n\n## Review Checklist\n\n### Standard (15 categories, \u002F15 scoring)\n\n| # | Category | What to check |\n|---|----------|--------------|\n| 1 | Installation | agents 25+, scripts 6+, skills 5+, rules 3, spec\u002Frules 8+, CLAUDE.md |\n| 2 | Settings & Hooks | env keys, SessionStart(2), PreToolUse(2), PostToolUse(3), Stop(1) |\n| 3 | Skill Selection | core skills vs package.json library matching |\n| 4 | Spec Quality | frontmatter (mock, testing, deps), REQ-NNN:, all required sections |\n| 5 | Design Quality | Components, State, Data Flow, Technical Decisions |\n| 6 | PLAN.md Quality | domain tags, model, wave, MSW tasks, Approved-at, Budget, Team Composition (team mode) |\n| 7 | TDD\u002FMock | MSW handlers for all endpoints, mock:true enforcement, test files |\n| 8 | Token Efficiency | file sizes \u003C200 lines, artifact limits, skill budget rules |\n| 9 | Code Quality | try\u002Fcatch, Zod, strict TS, no stubs, Server\u002FClient |\n| 10 | Flow Completeness | STATE.md, CONTEXT.md, [x] tasks, TEST_LOG.md |\n| 11 | Rules Size | all spec\u002Frules\u002F \u003C200 lines, no redundancy |\n| 12 | Skill Accuracy | manifest vs detected libs, on-demand coverage |\n| 13 | Extra Agents | security\u002Fa11y\u002Fperformance\u002Freview\u002Floop results |\n| 14 | Document Sync | CONTEXT.md consistent after modifications, no stale decisions |\n| 15 | Code Quality Detail | error handling grade, composition, naming, DRY |\n\n### Install Test Extended (20 categories, \u002F20 scoring)\n\nAll 15 standard categories plus:\n\n| # | Category | What to check |\n|---|----------|--------------|\n| 16 | Init Completeness | PROJECT.md framework detection, ARCHITECTURE.md structure, STATE.md features |\n| 17 | Security & Log Setup | SECURITY_STRATEGY.md, LOG_STRATEGY.md, audit findings quality |\n| 18 | Git Workflow | \u002Fcommit conventional format, REQ linking, no git add -A |\n| 19 | Debug & Loop | DEBUG.md hypotheses, LOOP_NOTES.md tracking, \u002Fstatus read-only |\n| 20 | Install Method Fidelity | npx: local files + manifest + doctor; marketplace: namespace + plugin paths |\n\n## Report Output\n\nTest artifacts are written to `__plugin-tests__\u002F` during execution:\n- Per-project: `TEST_LOG.md`, `REVIEW.md`\n- Extra agents: `SECURITY_REVIEW.md`, `ACCESSIBILITY_REVIEW.md`, `PERFORMANCE_REVIEW.md`, `REVIEW_RESULT.md`\n\nAfter each test run, generate a **key-summary English report** and save it to `reports\u002FREPORT_YYYYMMDD_HHMM.md` (inside this skill directory). The report should contain:\n- Date, branch, version, scope\n- Score summary table\n- Key findings (bullet points, not full prose)\n- Critical issues with severity\n- Recommendations\n- For `--install-test`: installation method comparison table + non-spec-dev feature results\n\nKeep each report under 80 lines. This is the **archival summary** — the raw artifacts in `__plugin-tests__\u002F` have full details.\n\n## Past Reports\n\nReports are stored in `reports\u002F` directory. **By default, read only the latest report** (highest datetime in filename) to understand current plugin quality baseline. Read older reports only when the user explicitly asks for trend analysis or historical comparison.\n\n| Report | Scope | Score |\n| --- | --- | --- |\n| `REPORT_20260321_1900.md` | Main branch validation (2 projects) | 9.5\u002F10 avg |\n| `REPORT_20260321_1700.md` | R5: Incremental fix-and-retest (3 projects) | 8.7\u002F10 avg |\n| `REPORT_20260321_1500.md` | R4: Full suite Phase 1+2 (4+2 projects, 15 categories) | 12.0\u002F15 avg |\n| `REPORT_20260321_1300.md` | R3: Monorepo solo vs team (10 features × 2 modes) | Solo 6\u002F10, Team 5\u002F10 |\n| `REPORT_20260321_1100.md` | R2: 4 projects × 4 features (post-fix) | 9.0\u002F10 avg |\n| `REPORT_20260321_0900.md` | R1: 4 projects × 1 feature (baseline) | 7.5\u002F10 avg |\n\n## NCC Source Path\n\nFor dev\u002FPhase 1\u002FPhase 2 modes: `node {NCC_SOURCE}\u002Fdist\u002Findex.js` (tests current branch build).\nFor install-test mode: uses published npm version (`npx nextjs-claude-code@latest`) and marketplace.\n","# \u002Fplugin-test — NCC Plugin E2E Test Suite\n\n## Usage\n\n```\n\u002Fplugin-test                    # Interactive — suggest a test mode based on recent changes\n\u002Fplugin-test --quick            # Phase 1 only: 4 diverse projects × 4 features each\n\u002Fplugin-test --full             # Phase 1 + Phase 2: full suite including monorepo + cross-feature mods\n\u002Fplugin-test --install-test     # Install test: 2 projects (npx + marketplace), 5 features each, non-spec-dev focus\n\u002Fplugin-test --phase1           # Phase 1 only (same as --quick)\n\u002Fplugin-test --phase2           # Phase 2 only: 10-feature monorepo (solo + team)\n\u002Fplugin-test --security         # Security-focused: run security-reviewer on all generated code\n\u002Fplugin-test --a11y             # Accessibility-focused: run browser-tester a11y checks on UI projects\n\u002Fplugin-test --codequality      # Code quality audit: deep analysis of all generated source files\n\u002Fplugin-test --token             # Token optimization audit: run token-optimizer on plugin source\n\u002Fplugin-test healthcare,marketplace   # Run specific projects only\n\u002Fplugin-test --create              # \u002Fcreate pipeline test: new + existing project ideation\n```\n\n## Test Modes\n\n### Phase 1: Quick Validation (--quick, --phase1)\n\n4 diverse projects, each with 4 features + extra agent verification.\n\n| Project | Stack | Extra Agents |\n|---------|-------|-------------|\n| `healthcare` | Prisma + next-auth + zod + vitest | `\u002Fsecurity`, `\u002Freview` |\n| `design-system` | framer-motion + playwright + storybook | `browser-tester`, `\u002Fqa --a11y` |\n| `realtime-collab` | drizzle + tanstack-query + zustand + msw | `performance-optimizer`, `\u002Floop` |\n| `marketplace` | Prisma + stripe + ai-sdk + @vercel\u002Fblob | `\u002Fsecurity --diff`, `\u002Freview`, `\u002Fstatus` |\n\nPer-project: NCC install → \u002Finit → 4× spec→dev (3 tasks each) → extra agent runs → TEST_LOG.md\n\nReviewer scores 15 categories:\n1-10: Installation, Hooks, Skills, Spec, Design, PLAN, TDD, Token, Code, Flow\n11: Rules size (\u003C200 lines each)\n12: Skill selection accuracy (manifest vs package.json)\n13: Extra agent execution results\n14: Document sync (CONTEXT.md\u002Fdesign.md consistency)\n15: Code quality detail (error handling, Zod, composition, Server\u002FClient)\n\n### Install Test (--install-test)\n\nTests the **published version** of NCC using two real installation methods. Primary focus: **non-spec-dev features**.\n\n| Project | Install Method | Stack | Features | Non-Spec-Dev Focus |\n|---------|---------------|-------|----------|-------------------|\n| `npx-healthcare` | `npx nextjs-claude-code@latest` | Prisma + next-auth + zod + vitest | 5 features | \u002Finit, \u002Fsecurity, \u002Flog, \u002Fcommit, \u002Frule, \u002Ftest, hooks, skills |\n| `mkt-marketplace` | `\u002Fplugin marketplace add` + `\u002Fplugin install` | Prisma + stripe + ai-sdk + @vercel\u002Fblob | 5 features | \u002Finit, \u002Freview, \u002Fstatus, \u002Fdebug, \u002Fqa, \u002Floop, \u002Fcommit, hooks, skills |\n\nPer-project: Install → \u002Finit → 5× spec→dev (2-3 tasks each) → non-spec-dev tests → TEST_LOG.md\n\nReviewer scores 20 categories (base 15 + 5 install-test categories):\n16: Init Completeness (PROJECT.md, ARCHITECTURE.md, STATE.md quality)\n17: Security & Log Setup (SECURITY_STRATEGY.md, LOG_STRATEGY.md, audit results)\n18: Git Workflow (\u002Fcommit conventional format + REQ linking)\n19: Debug & Loop (DEBUG.md hypothesis-driven, LOOP_NOTES.md tracking, \u002Fstatus read-only)\n20: Installation Method Fidelity (npx: local files + manifest; marketplace: namespace + CLAUDE_PLUGIN_ROOT)\n\n### Phase 2: Large-scale Monorepo (--phase2)\n\n2 identical CRM projects (solo vs team mode), each with:\n- 10 features in dependency chain\n- 5 cross-feature modifications (api-layer error format, RBAC expansion, tagging, cursor pagination, custom stages)\n- Document Sync verification (CONTEXT.md, design.md, history entries)\n- Solo vs Team comparison\n\nFeatures: api-layer → auth-system → contact-management → deal-pipeline → email-integration → activity-feed → reporting → team-workspace → notification-center → admin-settings\n\n### \u002Fcreate Pipeline Test (--create)\n\nTests the `\u002Fcreate` ideation pipeline on both new and existing projects.\n\n| Project | Scenario | Validation |\n|---------|----------|-----------|\n| `create-new` | Empty project → \u002Finit → \u002Fcreate \"AI recipe app\" | VISION.md, C-REVIEW.md, DECISION.md exist in spec\u002Fcreate\u002F |\n| `create-existing` | Healthcare project → \u002Fcreate \"notification system\" | Reads spec\u002FPROJECT.md context, outputs to spec\u002Fcreate\u002F |\n\nPer-project: NCC install → \u002Finit → \u002Fcreate → verify output docs → token isolation check → spec conversion test\n\nReviewer checks 5 categories:\n1. Document Quality: VISION.md has Problem\u002FDemand\u002FTarget\u002FValue sections\n2. C-Level Coverage: C-REVIEW.md has all 5 assessments (CEO\u002FCTO\u002FCPO\u002FCMO\u002FCDO) with verdicts\n3. Decision Clarity: DECISION.md has chosen approach + key decisions\n4. Token Isolation: After \u002Fcreate, run \u002Fspec and verify spec-writer does NOT read spec\u002Fcreate\u002F\n5. Spec Conversion: If converted, output is valid \u002Fspec command format\n\n### Full Suite (--full)\n\nRuns Phase 1, analyzes issues, then runs Phase 2. Generates consolidated REPORT with cross-phase comparison.\n\n### Security Audit (--security)\n\nFor each generated project:\n- Read installed `security-reviewer` agent\n- Scan all API routes for OWASP Top 10: injection, auth bypass, sensitive data exposure, CSRF\n- Check MSW handlers for auth mock bypass patterns\n- Check Zod validation coverage\n- Output: `SECURITY_REVIEW.md` per project\n\n### Accessibility Audit (--a11y)\n\nFor UI-heavy projects (design-system or any with UI components):\n- Read installed `browser-tester` agent\n- WCAG 2.1 AA checklist per component\n- Check: ARIA attributes, keyboard nav, focus management, color contrast\n- Check: prefers-reduced-motion respect, semantic HTML, role attributes\n- Output: `ACCESSIBILITY_REVIEW.md` per project\n\n### Code Quality Audit (--codequality)\n\nDeep source code analysis across all projects:\n- Error handling grade (try\u002Fcatch, error classification, no silent swallowing)\n- Zod validation grade (boundary validation, schema completeness)\n- TypeScript strictness (no any, no as unknown, proper inference)\n- Composition patterns (Context vs prop drilling, compound components)\n- Server\u002FClient discipline ('use client' minimization)\n- DRY compliance (no duplicate logic across routes)\n- Named constants (no magic numbers\u002Fstrings)\n- Output: code quality table in REPORT\n\n## Review Checklist\n\n### Standard (15 categories, \u002F15 scoring)\n\n| # | Category | What to check |\n|---|----------|--------------|\n| 1 | Installation | agents 25+, scripts 6+, skills 5+, rules 3, spec\u002Frules 8+, CLAUDE.md |\n| 2 | Settings & Hooks | env keys, SessionStart(2), PreToolUse(2), PostToolUse(3), Stop(1) |\n| 3 | Skill Selection | core skills vs package.json library matching |\n| 4 | Spec Quality | frontmatter (mock, testing, deps), REQ-NNN:, all required sections |\n| 5 | Design Quality | Components, State, Data Flow, Technical Decisions |\n| 6 | PLAN.md Quality | domain tags, model, wave, MSW tasks, Approved-at, Budget, Team Composition (team mode) |\n| 7 | TDD\u002FMock | MSW handlers for all endpoints, mock:true enforcement, test files |\n| 8 | Token Efficiency | file sizes \u003C200 lines, artifact limits, skill budget rules |\n| 9 | Code Quality | try\u002Fcatch, Zod, strict TS, no stubs, Server\u002FClient |\n| 10 | Flow Completeness | STATE.md, CONTEXT.md, [x] tasks, TEST_LOG.md |\n| 11 | Rules Size | all spec\u002Frules\u002F \u003C200 lines, no redundancy |\n| 12 | Skill Accuracy | manifest vs detected libs, on-demand coverage |\n| 13 | Extra Agents | security\u002Fa11y\u002Fperformance\u002Freview\u002Floop results |\n| 14 | Document Sync | CONTEXT.md consistent after modifications, no stale decisions |\n| 15 | Code Quality Detail | error handling grade, composition, naming, DRY |\n\n### Install Test Extended (20 categories, \u002F20 scoring)\n\nAll 15 standard categories plus:\n\n| # | Category | What to check |\n|---|----------|--------------|\n| 16 | Init Completeness | PROJECT.md framework detection, ARCHITECTURE.md structure, STATE.md features |\n| 17 | Security & Log Setup | SECURITY_STRATEGY.md, LOG_STRATEGY.md, audit findings quality |\n| 18 | Git Workflow | \u002Fcommit conventional format, REQ linking, no git add -A |\n| 19 | Debug & Loop | DEBUG.md hypotheses, LOOP_NOTES.md tracking, \u002Fstatus read-only |\n| 20 | Install Method Fidelity | npx: local files + manifest + doctor; marketplace: namespace + plugin paths |\n\n## Report Output\n\nTest artifacts are written to `__plugin-tests__\u002F` during execution:\n- Per-project: `TEST_LOG.md`, `REVIEW.md`\n- Extra agents: `SECURITY_REVIEW.md`, `ACCESSIBILITY_REVIEW.md`, `PERFORMANCE_REVIEW.md`, `REVIEW_RESULT.md`\n\nAfter each test run, generate a **key-summary English report** and save it to `reports\u002FREPORT_YYYYMMDD_HHMM.md` (inside this skill directory). The report should contain:\n- Date, branch, version, scope\n- Score summary table\n- Key findings (bullet points, not full prose)\n- Critical issues with severity\n- Recommendations\n- For `--install-test`: installation method comparison table + non-spec-dev feature results\n\nKeep each report under 80 lines. This is the **archival summary** — the raw artifacts in `__plugin-tests__\u002F` have full details.\n\n## Past Reports\n\nReports are stored in `reports\u002F` directory. **By default, read only the latest report** (highest datetime in filename) to understand current plugin quality baseline. Read older reports only when the user explicitly asks for trend analysis or historical comparison.\n\n| Report | Scope | Score |\n| --- | --- | --- |\n| `REPORT_20260321_1900.md` | Main branch validation (2 projects) | 9.5\u002F10 avg |\n| `REPORT_20260321_1700.md` | R5: Incremental fix-and-retest (3 projects) | 8.7\u002F10 avg |\n| `REPORT_20260321_1500.md` | R4: Full suite Phase 1+2 (4+2 projects, 15 categories) | 12.0\u002F15 avg |\n| `REPORT_20260321_1300.md` | R3: Monorepo solo vs team (10 features × 2 modes) | Solo 6\u002F10, Team 5\u002F10 |\n| `REPORT_20260321_1100.md` | R2: 4 projects × 4 features (post-fix) | 9.0\u002F10 avg |\n| `REPORT_20260321_0900.md` | R1: 4 projects × 1 feature (baseline) | 7.5\u002F10 avg |\n\n## NCC Source Path\n\nFor dev\u002FPhase 1\u002FPhase 2 modes: `node {NCC_SOURCE}\u002Fdist\u002Findex.js` (tests current branch build).\nFor install-test mode: uses published npm version (`npx nextjs-claude-code@latest`) and marketplace.\n",{"name":5,"agent":49,"context":50,"description":7,"argument-hint":51},"plugin-test-orchestrator","fork","--quick | --full | --install-test | --phase1 | --phase2 | --security | --a11y | --codequality | [project-name,...] | (no args = suggest mode)","main","330a71e3d12264515ce3477e446d29fd34a88a9d",".claude\u002Fskills\u002Fplugin-test",1,"https:\u002F\u002Fgithub.com\u002FByeongminLee\u002Fnextjs-claude-code","https:\u002F\u002Fbyeongminlee.github.io\u002Fnextjs-claude-code\u002F","https:\u002F\u002Fgithub.com\u002FByeongminLee","byeongmin.dev","2026-03-27 12:45:44",{"source":8,"skillName":5,"commands":62,"prompt":72,"zipAvailable":73},[63,66,69],{"tool":64,"command":65},"npx","npx skills add ByeongminLee\u002Fnextjs-claude-code --skill plugin-test",{"tool":67,"command":68},"bunx","bunx skills add ByeongminLee\u002Fnextjs-claude-code --skill plugin-test",{"tool":70,"command":71},"pnpm","pnpm dlx skills add ByeongminLee\u002Fnextjs-claude-code --skill plugin-test","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002FByeongminLee\u002Fnextjs-claude-code\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Fplugin-test\u002FSKILL.md",true,[75,81,85,89,93,97,101],{"path":76,"resourceType":77,"resourceTypeDesc":78,"sizeBytes":79,"sourceHtmlUrl":80},"reports\u002FREPORT_20260321_0900.md","REFERENCE","参考文档",1538,"https:\u002F\u002Fgithub.com\u002FByeongminLee\u002Fnextjs-claude-code\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Fplugin-test\u002Freports\u002FREPORT_20260321_0900.md",{"path":82,"resourceType":77,"resourceTypeDesc":78,"sizeBytes":83,"sourceHtmlUrl":84},"reports\u002FREPORT_20260321_1100.md",1148,"https:\u002F\u002Fgithub.com\u002FByeongminLee\u002Fnextjs-claude-code\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Fplugin-test\u002Freports\u002FREPORT_20260321_1100.md",{"path":86,"resourceType":77,"resourceTypeDesc":78,"sizeBytes":87,"sourceHtmlUrl":88},"reports\u002FREPORT_20260321_1300.md",1396,"https:\u002F\u002Fgithub.com\u002FByeongminLee\u002Fnextjs-claude-code\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Fplugin-test\u002Freports\u002FREPORT_20260321_1300.md",{"path":90,"resourceType":77,"resourceTypeDesc":78,"sizeBytes":91,"sourceHtmlUrl":92},"reports\u002FREPORT_20260321_1500.md",1306,"https:\u002F\u002Fgithub.com\u002FByeongminLee\u002Fnextjs-claude-code\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Fplugin-test\u002Freports\u002FREPORT_20260321_1500.md",{"path":94,"resourceType":77,"resourceTypeDesc":78,"sizeBytes":95,"sourceHtmlUrl":96},"reports\u002FREPORT_20260321_1700.md",1498,"https:\u002F\u002Fgithub.com\u002FByeongminLee\u002Fnextjs-claude-code\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Fplugin-test\u002Freports\u002FREPORT_20260321_1700.md",{"path":98,"resourceType":77,"resourceTypeDesc":78,"sizeBytes":99,"sourceHtmlUrl":100},"reports\u002FREPORT_20260321_1900.md",1529,"https:\u002F\u002Fgithub.com\u002FByeongminLee\u002Fnextjs-claude-code\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Fplugin-test\u002Freports\u002FREPORT_20260321_1900.md",{"path":102,"resourceType":77,"resourceTypeDesc":78,"sizeBytes":103,"sourceHtmlUrl":104},"reports\u002FREPORT_20260322_1400.md",2943,"https:\u002F\u002Fgithub.com\u002FByeongminLee\u002Fnextjs-claude-code\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Fplugin-test\u002Freports\u002FREPORT_20260322_1400.md",[106,124,132,162,190,228],{"skillNo":107,"name":108,"nameZh":6,"description":109,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":110,"categories":111,"keywords":112,"publishedAt":44},"01M6E263DRDCZ5AX01N6MDYC6Z","zustand","Zustand state management guide. Use when working with store code (src\u002Fstore\u002F**), implementing actions, managing state, or creating slices. Triggers on Zustand store development, state management questions, or action implementation.","https:\u002F\u002Fgithub.com\u002FByeongminLee\u002Fnextjs-claude-code\u002Fblob\u002Fmain\u002Ftemplate\u002Fskills-archive\u002Fzustand\u002FSKILL.md",[],[108,113,114,115,116,117,118,119,120,121,122,41,123],"triggers","state management","zustand state","implementation","use","guide","management","src","actions","state","development",{"skillNo":125,"name":126,"nameZh":6,"description":6,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":127,"categories":128,"keywords":129,"publishedAt":44},"01M6E260QFP3ZMXY2Y6KTGZYXS","web-vitals","https:\u002F\u002Fgithub.com\u002FByeongminLee\u002Fnextjs-claude-code\u002Fblob\u002Fmain\u002Ftemplate\u002Fskills-archive\u002Fweb-vitals\u002FSKILL.md",[],[130,131],"web vitals","web",{"skillNo":133,"name":134,"nameZh":6,"description":135,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":136,"categories":137,"keywords":141,"publishedAt":44},"01M6E25QEKXHH092S5RB0GKM0K","vercel-storage","Vercel storage expert guidance — Blob, Edge Config, and Marketplace storage (Neon Postgres, Upstash Redis). Use when choosing, configuring, or using data storage with Vercel applications.","https:\u002F\u002Fgithub.com\u002FByeongminLee\u002Fnextjs-claude-code\u002Fblob\u002Fmain\u002Ftemplate\u002Fskills-archive\u002Fvercel-storage\u002FSKILL.md",[138],{"categoryNo":139,"categoryName":140,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000010","后端与接口",[142,143,144,145,146,147,148,149,150,151,152,153,154,155,156,157,158,159,160,161],"vercel","vercel storage","storage","supabase","api","mongodb","SET","nextjs","database","auth","migrate","click","app","neon","prisma","postgres","expert","up","auto","configure",{"skillNo":163,"name":164,"nameZh":6,"description":165,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":166,"categories":167,"keywords":174,"publishedAt":44},"01M6E25EC9YM4QM0KBQ2W5R1SC","vercel-functions","Vercel Functions expert guidance — Serverless Functions, Edge Functions, Fluid Compute, streaming, Cron Jobs, and runtime configuration. Use when configuring, debugging, or optimizing server-side code running on Vercel.","https:\u002F\u002Fgithub.com\u002FByeongminLee\u002Fnextjs-claude-code\u002Fblob\u002Fmain\u002Ftemplate\u002Fskills-archive\u002Fvercel-functions\u002FSKILL.md",[168,169],{"categoryNo":139,"categoryName":140,"taxonomyType":21,"taxonomyTypeDesc":22},{"categoryNo":170,"categoryName":171,"taxonomyType":172,"taxonomyTypeDesc":173},"01JBSEEDCAT000000000000030","开发工程师","PROFESSION","适用职业",[175,142,176,177,178,31,179,180,154,155,181,182,183,184,185,186,187,188,189,158],"functions","vercel functions","local","shared","time","apps","express","vercel ai sdk","router","loop","server","logs","streaming","delete","create",{"skillNo":191,"name":192,"nameZh":6,"description":193,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":194,"categories":195,"keywords":211,"publishedAt":44},"01M6E25873YRYP96MSQK7E2D9X","ui-ux-pro-max","UI\u002FUX design intelligence for web and mobile. Includes 50+ styles, 161 color palettes, 57 font pairings, 161 product types, 99 UX guidelines, and 25 chart types across 10 stacks (React, Next.js, Vue, Svelte, SwiftUI, React Native, Flutter, Tailwind, shadcn\u002Fui, and HTML\u002FCSS). Actions: plan, build, create, design, implement, review, fix, improve, optimize, enhance, refactor, and check UI\u002FUX code. Projects: website, landing page, dashboard, admin panel, e-commerce, SaaS, portfolio, blog, and mobile app. Elements: button, modal, navbar, sidebar, card, table, form, and chart. Styles: glassmorphism, claymorphism, minimalism, brutalism, neumorphism, bento grid, dark mode, responsive, skeuomorphism, and flat design. Topics: color systems, accessibility, animation, layout, typography, font pairing, spacing, interaction states, shadow, and gradient. Integrations: shadcn\u002Fui MCP for component search and examples.","https:\u002F\u002Fgithub.com\u002FByeongminLee\u002Fnextjs-claude-code\u002Fblob\u002Fmain\u002Ftemplate\u002Fskills-archive\u002Fui-ux-pro-max\u002FSKILL.md",[196,199,202,205,208],{"categoryNo":197,"categoryName":198,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000009","前端开发",{"categoryNo":200,"categoryName":201,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000015","视觉设计",{"categoryNo":203,"categoryName":204,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000023","Skill与智能体开发",{"categoryNo":206,"categoryName":207,"taxonomyType":172,"taxonomyTypeDesc":173},"01JBSEEDCAT000000000000032","设计师",{"categoryNo":209,"categoryName":210,"taxonomyType":172,"taxonomyTypeDesc":173},"01JBSEEDCAT000000000000033","数据分析师",[212,213,214,215,216,217,218,219,154,220,221,222,38,223,224,225,226,121,227,189],"ui ux pro max","ui","ux","max","css","MCP","design","react","card","chart","fix","plan","improve","Tailwind","Animation","search",{"skillNo":229,"name":230,"nameZh":6,"description":231,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":232,"categories":233,"keywords":234,"publishedAt":44},"01M6E251G3FJSRB9019V20QK9X","turborepo","Turborepo monorepo build system guidance. Triggers on: turbo.json, task pipelines,\ndependsOn, caching, remote cache, the \"turbo\" CLI, --filter, --affected, CI optimization, environment\nvariables, internal packages, monorepo structure\u002Fbest practices, and boundaries.\n\nUse when user: configures tasks\u002Fworkflows\u002Fpipelines, creates packages, sets up\nmonorepo, shares code between apps, runs changed\u002Faffected packages, debugs cache,\nor has apps\u002Fpackages directories.","https:\u002F\u002Fgithub.com\u002FByeongminLee\u002Fnextjs-claude-code\u002Fblob\u002Fmain\u002Ftemplate\u002Fskills-archive\u002Fturborepo\u002FSKILL.md",[],[230,235,117,236,237,180,113,238,239,159,240,37,241,242,243,244,41,245,246,247],"cli","filter","environment variables","caching","structure","environment","system","canary","Optimization","ci","packages","tasks","workflows"]