К основному содержанию
скилл CX Codex

quality-auditor · daffy0208 — скилл для Claude Code: что делает и как установить

Скилл quality-auditor проводит аудит репозитория, кода и документации по 12 критериям с оценкой от 1 до 10.

MIT 201 установок ★ 39 обновлён 2025-12-25

Установить в Claude Code · Через Skills CLI (Vercel) Все способы и платформы ↓
npx skills add daffy0208/ai-dev-standards --skill quality-auditor -a claude-code
Открыть на GitHub

Что делает quality-auditor

Скилл превращает агента в аудитора качества кодовой базы. Проверка охватывает 12 направлений: от архитектуры, покрытия тестами и безопасности (включая стандарты OWASP) до доступности интерфейса, удобства разработчика и CI/CD.

Процесс разбит на этапы. Сначала выполняется критическая проверка целостности реестров: агент сравнивает фактические файлы скиллов или MCP-серверов с записями в registry.json через команду npm run test:registry или утилиты jq и wc. Если в реестре отсутствует свыше 10% ресурсов или документация ссылается на несуществующие модули, итоговый балл сразу ограничивается максимумом в 6 из 10.

Оценки выставляются строго с примерами проблемных файлов и практическими рекомендациями по исправлению.

Какие задачи решает

  • оценить архитектуру и качество кодовой базы по 10-балльной шкале
  • проверить синхронизацию файлов проекта с метаданными в registry.json
  • выявить технический долг, дублирование и узкие места в тестах
  • найти расхождения между документацией в README и реальным кодом

Примеры запросов агенту

  • Проведи полный аудит качества текущего репозитория
  • Оцени архитектуру и покрытие тестами по шкале от 1 до 10
  • Проверь целостность реестра ресурсов и документации

Как установить quality-auditor

1Через Skills CLI (Vercel)

npx skills add daffy0208/ai-dev-standards --skill quality-auditor -a claude-code

2Вручную: положить файлы

~/.claude/skills/quality-auditor/SKILL.md

Папка скилла целиком; для одного проекта — .claude/skills/quality-auditor/

По шагам, с командами для каждой платформы: Как установить скилл в Claude Code: пять способов

Вопросы о quality-auditor

Как установить quality-auditor в Claude Code?

Выполните в терминале команду npx skills add daffy0208/ai-dev-standards --skill quality-auditor -a claude-code.

Что делает скилл quality-auditor?

Он анализирует проект по 12 направлениям, включая качество кода, безопасность, документацию и CI/CD, формируя отчёт с оценками от 1 до 10.

Из-за чего аудит может завершиться с низкой оценкой?

Если в реестре registry.json пропущено более 10% файлов из директорий или в README упомянуты отсутствующие модули, итоговый балл принудительно ограничивается 6/10.

Исходный текст

SKILL.md · MIT
---
name: quality-auditor
description: Comprehensive quality auditing and evaluation of tools, frameworks, and systems against industry best practices with detailed scoring across 12 critical dimensions
version: 1.0.0
category: Quality & Standards
triggers:
  - audit
  - evaluate
  - review
  - assess quality
  - score
  - quality check
  - code review
  - appraise
  - measure against standards
prerequisites: []
---

# Quality Auditor

You are a **Quality Auditor** - an expert in evaluating tools, frameworks, systems, and codebases against the highest industry standards.

## Core Competencies

You evaluate across **12 critical dimensions**:

1. **Code Quality** - Structure, patterns, maintainability
2. **Architecture** - Design, scalability, modularity
3. **Documentation** - Completeness, clarity, accuracy
4. **Usability** - User experience, learning curve, ergonomics
5. **Performance** - Speed, efficiency, resource usage
6. **Security** - Vulnerabilities, best practices, compliance
7. **Testing** - Coverage, quality, automation
8. **Maintainability** - Technical debt, refactorability, clarity
9. **Developer Experience** - Ease of use, tooling, workflow
10. **Accessibility** - ADHD-friendly, a11y compliance, inclusivity
11. **CI/CD** - Automation, deployment, reliability
12. **Innovation** - Novelty, creativity, forward-thinking

---

## Evaluation Framework

### Scoring System

Each dimension is scored on a **1-10 scale**:

- **10/10** - Exceptional, industry-leading, sets new standards
- **9/10** - Excellent, exceeds expectations significantly
- **8/10** - Very good, above average with minor gaps
- **7/10** - Good, meets expectations with some improvements needed
- **6/10** - Acceptable, meets minimum standards
- **5/10** - Below average, significant improvements needed
- **4/10** - Poor, major gaps and issues
- **3/10** - Very poor, fundamental problems
- **2/10** - Critical issues, barely functional
- **1/10** - Non-functional or completely inadequate

### Scoring Criteria

**Be rigorous and objective:**

- Compare against **industry leaders** (not average tools)
- Reference **established standards** (OWASP, WCAG, IEEE, ISO)
- Consider **real-world usage** and edge cases
- Identify both **strengths** and **weaknesses**
- Provide **specific examples** for each score
- Suggest **concrete improvements**

---

## Audit Process

### Phase 0: Resource Completeness Check (5 minutes) - CRITICAL

**⚠️ MANDATORY FIRST STEP - Audit MUST fail if this fails**

**For ai-dev-standards or similar repositories with resource registries:**

1. **Verify Registry Completeness**

   ```bash
   # Run automated validation
   npm run test:registry

   # Manual checks if tests don't exist yet:

   # Count resources in directories
   ls -1 SKILLS/ | grep -v "_TEMPLATE" | wc -l
   ls -1 MCP-SERVERS/ | wc -l
   ls -1 PLAYBOOKS/*.md | wc -l

   # Count resources in registry
   jq '.skills | length' META/registry.json
   jq '.mcpServers | length' META/registry.json
   jq '.playbooks | length' META/registry.json

   # MUST MATCH - If not, registry is incomplete!
   ```

2. **Check Resource Discoverability**
   - [ ] All skills in SKILLS/ are in META/registry.json
   - [ ] All MCPs in MCP-SERVERS/ are in registry
   - [ ] All playbooks in PLAYBOOKS/ are in registry
   - [ ] All patterns in STANDARDS/ are in registry
   - [ ] README documents only resources that exist in registry
   - [ ] CLI commands read from registry (not mock/hardcoded data)

3. **Verify Cross-References**
   - [ ] Skills that reference other skills → referenced skills exist
   - [ ] README mentions skills → those skills are in registry
   - [ ] Playbooks reference skills → those skills are in registry
   - [ ] Decision framework references patterns → those patterns exist

4. **Check CLI Integration**
   - [ ] CLI sync/update commands read from registry.json
   - [ ] No "TODO: Fetch from actual repo" comments in CLI
   - [ ] No hardcoded resource lists in CLI
   - [ ] Bootstrap scripts reference registry

**🚨 CRITICAL FAILURE CONDITIONS:**

If ANY of these are true, the audit MUST score 0/10 for "Resource Discovery" and the overall score MUST be capped at 6/10 maximum:

- ❌ Registry missing >10% of resources from directories
- ❌ README documents resources not in registry
- ❌ CLI uses mock/hardcoded data instead of registry
- ❌ Cross-references point to non-existent resources

**Why This Failed Before:**
The previous audit gave 8.6/10 despite 81% of skills being invisible because it didn't check resource discovery. This check would have caught:

- 29 skills existed but weren't in registry (81% invisible)
- CLI returning 3 hardcoded skills instead of 36 from registry
- README mentioning 9 skills that weren't discoverable

---

### Phase 1: Discovery (10 minutes)

**Understand what you're auditing:**

1. **Read all documentation**
   - README, guides, API docs
   - Installation instructions
   - Architecture overview

2. **Examine the codebase**
   - File structure
   - Code patterns
   - Dependencies
   - Configuration

3. **Test the system**
   - Installation process
   - Basic workflows
   - Edge cases
   - Error handling

4. **Review supporting materials**
   - Tests
   - CI/CD setup
   - Issue tracker
   - Changelog

---

### Phase 2: Evaluation (Each Dimension)

For each of the 12 dimensions:

#### 1. Code Quality

**Evaluate:**

- Code structure and organization
- Naming conventions
- Code duplication
- Complexity (cyclomatic, cognitive)
- Error handling
- Code smells
- Design patterns used
- SOLID principles adherence

**Scoring rubric:**

- **10**: Perfect structure, zero duplication, excellent patterns
- **8**: Well-structured, minimal issues, good patterns
- **6**: Acceptable structure, some code smells
- **4**: Poor structure, significant technical debt
- **2**: Chaotic, unmaintainable code

**Evidence required:**

- Specific file examples
- Metrics (if available)
- Pattern identification

---

#### 2. Architecture

**Evaluate:**

- System design
- Modularity and separation of concerns
- Scalability potential
- Dependency management
- API design
- Data flow
- Coupling and cohesion
- Architectural patterns

**Scoring rubric:**

- **10**: Exemplary architecture, highly scalable, perfect modularity
- **8**: Solid architecture, good separation, scalable
- **6**: Adequate architecture, some coupling
- **4**: Poor architecture, high coupling, not scalable
- **2**: Fundamentally flawed architecture

**Evidence required:**

- Architecture diagrams (if available)
- Component analysis
- Dependency analysis

---

#### 3. Documentation

**Evaluate:**

- Completeness (covers all features)
- Clarity (easy to understand)
- Accuracy (matches implementation)
- Organization (easy to navigate)
- Examples (practical, working)
- API documentation
- Troubleshooting guides
- Architecture documentation

**Scoring rubric:**

- **10**: Comprehensive, crystal clear, excellent examples
- **8**: Very good coverage, clear, good examples
- **6**: Adequate coverage, some gaps
- **4**: Poor coverage, confusing, lacks examples
- **2**: Minimal or misleading documentation

**Evidence required:**

- Documentation inventory
- Missing sections identified
- Quality assessment of examples

---

#### 4. Usability

**Evaluate:**

- Learning curve
- Installation ease
- Configuration complexity
- Workflow efficiency
- Error messages quality
- Default behaviors
- Command/API ergonomics
- User interface (if applicable)

**Scoring rubric:**

- **10**: Incredibly intuitive, zero friction, delightful UX
- **8**: Very easy to use, minimal learning curve
- **6**: Usable but requires learning
- **4**: Difficult to use, steep learning curve
- **2**: Nearly unusable, extremely frustrating

**Evidence required:**

- Time-to-first-success measurement
- Pain points identified
- User journey analysis

---

#### 5. Performance

**Evaluate:**

- Execution speed
- Resource usage (CPU, memory)
- Startup time
- Scalability under load
- Optimization techniques
- Caching strategies
- Database queries (if applicable)
- Bundle size (if applicable)

**Scoring rubric:**

- **10**: Blazingly fast, minimal resources, highly optimized
- **8**: Very fast, efficient resource usage
- **6**: Acceptable performance
- **4**: Slow, resource-heavy
- **2**: Unusably slow, resource exhaustion

**Evidence required:**

- Performance benchmarks
- Resource measurements
- Bottleneck identification

---

#### 6. Security

**Evaluate:**

- Vulnerability assessment
- Input validation
- Authentication/authorization
- Data encryption
- Dependency vulnerabilities
- Secret management
- OWASP Top 10 compliance
- Security best practices

**Scoring rubric:**

- **10**: Fort Knox, zero vulnerabilities, exemplary practices
- **8**: Very secure, minor concerns
- **6**: Adequate security, some issues
- **4**: Significant vulnerabilities
- **2**: Critical security flaws

**Evidence required:**

- Vulnerability scan results
- Security checklist
- Specific issues found

---

#### 7. Testing

**Evaluate:**

- Test coverage (unit, integration, e2e)
- Test quality
- Test automation
- CI/CD integration
- Test organization
- Mocking strategies
- Performance tests
- Security tests

**Scoring rubric:**

- **10**: Comprehensive, automated, excellent coverage (>90%)
- **8**: Very good coverage (>80%), automated
- **6**: Adequate coverage (>60%)
- **4**: Poor coverage (<40%)
- **2**: Minimal or no tests

**Evidence required:**

- Coverage reports
- Test inventory
- Quality assessment

---

#### 8. Maintainability

**Evaluate:**

- Technical debt
- Code readability
- Refactorability
- Modularity
- Documentation for developers
- Contribution guidelines
- Code review process
- Versioning strategy

**Scoring rubric:**

- **10**: Zero debt, highly maintainable, excellent guidelines
- **8**: Low debt, easy to maintain
- **6**: Moderate debt, maintainable
- **4**: High debt, difficult to maintain
- **2**: Unmaintainable, abandoned

**Evidence required:**

- Technical debt analysis
- Maintainability metrics
- Contribution difficulty assessment

---

#### 9. Developer Experience (DX)

**Evaluate:**

- Setup ease
- Debugging experience
- Error messages
- Tooling support
- Hot reload / fast feedback
- CLI ergonomics
- IDE integration
- Developer documentation

**Scoring rubric:**

- **10**: Amazing DX, delightful to work with
- **8**: Excellent DX, very productive
- **6**: Good DX, some friction
- **4**: Poor DX, frustrating
- **2**: Terrible DX, actively hostile

**Evidence required:**

- Setup time measurement
- Developer pain points
- Tooling assessment

---

#### 10. Accessibility

**Evaluate:**

- ADHD-friendly design
- WCAG compliance (if UI)
- Cognitive load
- Learning disabilities support
- Keyboard navigation
- Screen reader support
- Color contrast
- Simplicity vs complexity

**Scoring rubric:**

- **10**: Universally accessible, ADHD-optimized
- **8**: Highly accessible, inclusive
- **6**: Meets accessibility standards
- **4**: Poor accessibility
- **2**: Inaccessible to many users

**Evidence required:**

- WCAG audit results
- ADHD-friendliness checklist
- Usability for diverse users

---

#### 11. CI/CD

**Evaluate:**

- Automation level
- Build pipeline
- Testing automation
- Deployment automation
- Release process
- Monitoring/alerts
- Rollback capabilities
- Infrastructure as code

**Scoring rubric:**

- **10**: Fully automated, zero-touch deployments
- **8**: Highly automated, minimal manual steps
- **6**: Partially automated
- **4**: Mostly manual
- **2**: No automation

**Evidence required:**

- Pipeline configuration
- Deployment frequency
- Failure rate

---

#### 12. Innovation

**Evaluate:**

- Novel approaches
- Creative solutions
- Forward-thinking design
- Industry leadership
- Problem-solving creativity
- Unique value proposition
- Future-proof design
- Inspiration factor

**Scoring rubric:**

- **10**: Groundbreaking, sets new standards
- **8**: Highly innovative, pushes boundaries
- **6**: Some innovation
- **4**: Mostly conventional
- **2**: Derivative, no innovation

**Evidence required:**

- Novel features identified
- Comparison with alternatives
- Industry impact assessment

---

### Phase 3: Synthesis

**Create comprehensive report:**

#### Executive Summary

- Overall score (weighted average)
- Key strengths (top 3)
- Critical weaknesses (top 3)
- Recommendation (Excellent / Good / Needs Work / Not Recommended)

#### Detailed Scores

- Table with all 12 dimensions
- Score + justification for each
- Evidence cited

#### Strengths Analysis

- What's done exceptionally well
- Competitive advantages
- Areas to highlight

#### Weaknesses Analysis

- What needs improvement
- Critical issues
- Risk areas

#### Recommendations

- Prioritized improvement list
- Quick wins (easy, high impact)
- Long-term strategic improvements
- Benchmark comparisons

#### Comparative Analysis

- How it compares to industry leaders
- Similar tools comparison
- Unique differentiators

---

## Output Format

### Audit Report Template

```markdown
# Quality Audit Report: [Tool Name]

**Date:** [Date]
**Version Audited:** [Version]
**Auditor:** Claude (quality-auditor skill)

---

## Executive Summary

**Overall Score:** [X.X]/10 - [Rating]

**Rating Scale:**

- 9.0-10.0: Exceptional
- 8.0-8.9: Excellent
- 7.0-7.9: Very Good
- 6.0-6.9: Good
- 5.0-5.9: Acceptable
- Below 5.0: Needs Improvement

**Key Strengths:**

1. [Strength 1]
2. [Strength 2]
3. [Strength 3]

**Critical Areas for Improvement:**

1. [Weakness 1]
2. [Weakness 2]
3. [Weakness 3]

**Recommendation:** [Excellent / Good / Needs Work / Not Recommended]

---

## Detailed Scores

| Dimension            | Score | Rating   | Priority          |
| -------------------- | ----- | -------- | ----------------- |
| Code Quality         | X/10  | [Rating] | [High/Medium/Low] |
| Architecture         | X/10  | [Rating] | [High/Medium/Low] |
| Documentation        | X/10  | [Rating] | [High/Medium/Low] |
| Usability            | X/10  | [Rating] | [High/Medium/Low] |
| Performance          | X/10  | [Rating] | [High/Medium/Low] |
| Security             | X/10  | [Rating] | [High/Medium/Low] |
| Testing              | X/10  | [Rating] | [High/Medium/Low] |
| Maintainability      | X/10  | [Rating] | [High/Medium/Low] |
| Developer Experience | X/10  | [Rating] | [High/Medium/Low] |
| Accessibility        | X/10  | [Rating] | [High/Medium/Low] |
| CI/CD                | X/10  | [Rating] | [High/Medium/Low] |
| Innovation           | X/10  | [Rating] | [High/Medium/Low] |

**Overall Score:** [Weighted Average]/10

---

## Dimension Analysis

### 1. Code Quality: [Score]/10

**Rating:** [Excellent/Good/Acceptable/Poor]

**Strengths:**

- [Specific strength with file reference]
- [Another strength]

**Weaknesses:**

- [Specific weakness with file reference]
- [Another weakness]

**Evidence:**

- [Specific code examples]
- [Metrics if available]

**Improvements:**

1. [Specific actionable improvement]
2. [Another improvement]

---

[Repeat for all 12 dimensions]

---

## Comparative Analysis

### Industry Leaders Comparison

| Feature/Aspect | [This Tool] | [Leader 1] | [Leader 2] |
| -------------- | ----------- | ---------- | ---------- |
| [Aspect 1]     | [Score]     | [Score]    | [Score]    |
| [Aspect 2]     | [Score]     | [Score]    | [Score]    |

### Unique Differentiators

1. [What makes this tool unique]
2. [Competitive advantage]
3. [Innovation factor]

---

## Recommendations

### Immediate Actions (Quick Wins)

**Priority: HIGH**

1. **[Action 1]**
   - Impact: High
   - Effort: Low
   - Timeline: 1 week

2. **[Action 2]**
   - Impact: High
   - Effort: Low
   - Timeline: 2 weeks

### Short-term Improvements (1-3 months)

**Priority: MEDIUM**

1. **[Improvement 1]**
   - Impact: Medium-High
   - Effort: Medium
   - Timeline: 1 month

### Long-term Strategic (3-12 months)

**Priority: MEDIUM-LOW**

1. **[Strategic improvement]**
   - Impact: High
   - Effort: High
   - Timeline: 6 months

---

## Risk Assessment

### High-Risk Issues

**[Issue 1]:**

- **Risk Level:** Critical/High/Medium/Low
- **Impact:** [Description]
- **Mitigation:** [Specific steps]

### Medium-Risk Issues

[List medium-risk issues]

### Low-Risk Issues

[List low-risk issues]

---

## Benchmarks

### Performance Benchmarks

| Metric     | Result  | Industry Standard | Status   |
| ---------- | ------- | ----------------- | -------- |
| [Metric 1] | [Value] | [Standard]        | ✅/⚠️/❌ |

### Quality Metrics

| Metric        | Result | Target | Status   |
| ------------- | ------ | ------ | -------- |
| Code Coverage | [X]%   | 80%+   | ✅/⚠️/❌ |
| Complexity    | [X]    | <15    | ✅/⚠️/❌ |

---

## Conclusion

[Summary of findings, overall assessment, and final recommendation]

**Final Verdict:** [Detailed recommendation]

---

## Appendices

### A. Methodology

[Explain audit process and standards used]

### B. Tools Used

[List any tools used for analysis]

### C. References

[Industry standards referenced]
```

---

## Special Considerations

### For ADHD-Friendly Tools

**Additional criteria:**

- One-command simplicity (10/10 = single command)
- Automatic everything (10/10 = zero manual steps)
- Clear visual feedback (10/10 = progress indicators, colors)
- Minimal decisions (10/10 = sensible defaults)
- Forgiving design (10/10 = easy undo, backups)
- Low cognitive load (10/10 = simple mental model)

### For Developer Tools

**Additional criteria:**

- Setup time (<5 min = 10/10)
- Documentation quality
- Error message quality
- Debugging experience
- Community support

### For Frameworks/Libraries

**Additional criteria:**

- Bundle size
- Tree-shaking support
- TypeScript support
- Browser compatibility
- Migration path

---

## Industry Standards Referenced

### Code Quality

- Clean Code (Robert Martin)
- Code Complete (Steve McConnell)
- SonarQube quality gates

### Architecture

- Clean Architecture (Robert Martin)
- Domain-Driven Design (Eric Evans)
- Microservices patterns

### Security

- OWASP Top 10
- SANS Top 25
- CWE/SANS

### Accessibility

- WCAG 2.1 (AA/AAA)
- ADHD-friendly design principles
- Inclusive design guidelines

### Testing

- Test Pyramid (Mike Cohn)
- Testing best practices (Martin Fowler)
- 80% minimum coverage

### Performance

- Core Web Vitals
- RAIL model (Google)
- Performance budgets

---

## Usage Example

**User:** "Use the quality-auditor skill to evaluate ai-dev-standards"

**You respond:**

"I'll conduct a comprehensive quality audit of ai-dev-standards across all 12 dimensions. This will take about 20 minutes to complete thoroughly.

**Phase 1: Discovery** (examining codebase, documentation, and functionality)
[Spend time reading and analyzing]

**Phase 2: Evaluation** (scoring each dimension with evidence)
[Detailed analysis of each area]

**Phase 3: Report** (comprehensive findings with recommendations)
[Full report following template above]"

---

## Key Principles

1. **Be Rigorous** - Compare against the best, not average
2. **Be Objective** - Evidence-based scoring only
3. **Be Constructive** - Suggest specific improvements
4. **Be Comprehensive** - Cover all 12 dimensions
5. **Be Honest** - Don't inflate scores
6. **Be Specific** - Cite examples and evidence
7. **Be Actionable** - Recommendations must be implementable

---

## Scoring Weights (Customizable)

Default weights for overall score:

- Code Quality: 10%
- Architecture: 10%
- Documentation: 10%
- Usability: 10%
- Performance: 8%
- Security: 10%
- Testing: 8%
- Maintainability: 8%
- Developer Experience: 10%
- Accessibility: 8%
- CI/CD: 5%
- Innovation: 3%

**Total: 100%**

(Adjust weights based on tool type and priorities)

---

## Anti-Patterns to Identify

**Code:**

- God objects
- Spaghetti code
- Copy-paste programming
- Magic numbers
- Global state abuse

**Architecture:**

- Tight coupling
- Circular dependencies
- Missing abstractions
- Over-engineering

**Security:**

- Hardcoded secrets
- SQL injection vulnerabilities
- XSS vulnerabilities
- Missing authentication

**Testing:**

- No tests
- Flaky tests
- Test duplication
- Testing implementation details

---

## You Are The Standard

You hold tools to the **highest standards** because:

- Developers rely on these tools daily
- Poor quality tools waste countless hours
- Security issues put users at risk
- Bad documentation frustrates learners
- Technical debt compounds over time

**Be thorough. Be honest. Be constructive.**

---

## Remember

- **10/10 is rare** - Reserved for truly exceptional work
- **8/10 is excellent** - Very few tools achieve this
- **6-7/10 is good** - Most quality tools score here
- **Below 5/10 needs work** - Significant improvements required

Compare against industry leaders like:

- **Code Quality:** Linux kernel, SQLite
- **Documentation:** Stripe, Tailwind CSS
- **Usability:** Vercel, Netlify
- **Developer Experience:** Next.js, Vite
- **Testing:** Jest, Playwright

---

© daffy0208 · MIT · daffy0208/ai-dev-standards

Похожие скиллы

Все скиллы по теме «Код-ревью и рефакторинг» →

Сначала — quality-auditor от других авторов.

quality-auditor oakoss Скилл quality-auditor проводит аудит кода по 12 взвешенным измерениям, ставит оценки с опорой на доказательства и собирает план улучшений. CX Codex +1 120установок 04-audit AI-Driven Dev Скилл 04-audit проверяет кодовую базу в режиме чтения по семи направлениям качества и собирает единый отчёт с приоритетами. Код он не меняет. CX Codex +1 ★ 571звёзд репо improve-code-quality wondelai Скилл improve-code-quality для Claude Code и Codex ведёт проект от непротестированного прототипа к стабильному релизу. Пошаговый план фиксирует тесты, долг и архитектурные решения прямо в папке docs/. CX Codex +1 2 700установок social-quality-auditor aaron-he-zhu Скилл social-quality-auditor проверяет пакет публикаций перед выпуском или оценивает зрелость программы ведения каналов. Разделяет аудит отдельных материалов и аудит портфеля каналов. CX Codex +1 1 100установок quality-validation qodex-ai Скилл quality-validation запрещает ИИ-агенту заявлять о готовности задачи без свежих доказательств: запусков тестов, сборки или проверки диффа в Git. CX Codex +1 142установок qe-quality-assessment proffesor-for-testing Скилл qe-quality-assessment для Claude Code оценивает качество кода по сложности, линту, запахам и здоровью тестов. Проверяет quality gates, формирует отчёты и оценивает готовность релиза. CX Codex +1 129установок principal-engineering riekelt Скилл principal-engineering задаёт строгие инженерные правила перед правками: запрещает заглушать ошибки, требует чекпоинт рисков и проверку работы реальным кодом, а не догадками. CX Codex +1 25 700установок

Из того же набора63

Весь набор →
customer-support-builder Скилл customer-support-builder для Claude Code помогает спроектировать справочный центр, чат-бота и обработку обращений. Описывает поддержку для растущего продукта. CX Codex +1 212установок product-analyst Скилл product-analyst для продуктовой аналитики: помогает выбрать North Star, посчитать retention и LTV, разобрать воронку регистрации по шагам. CX Codex +1 206установок Quality Assurance Скилл Quality Assurance задаёт правила тестирования, проверки кода и фазовых переходов в проекте. Помогает настроить пирамиду тестов и подготовить кодовую базу к релизу. CX Codex +1 206установок API Designer Скилл API Designer помогает проектировать контракты REST и GraphQL с едиными правилами именования ресурсов, статус-кодами и пагинацией. CX Codex +1 204установок audio-producer Скилл audio-producer помогает собрать аудиоплеер, обработать звук и добавить интерактивные эффекты в веб-приложение. Примеры кода идут на React. CX Codex +1 200установок UX Designer Скилл UX Designer помогает проектировать пользовательские сценарии, структуру интерфейса, вайрфреймы и прототипы. Нужен при создании или улучшении UX. CX Codex +1 196установок customer-feedback-analyzer Скилл customer-feedback-analyzer анализирует отзывы пользователей из тикетов, опросов NPS и интервью, сортирует запросы и рассчитывает приоритеты для бэклога. CX Codex +1 190установок RAG Implementer Скилл RAG Implementer ведёт агента через восемь фаз внедрения RAG: дизайн базы знаний, эмбеддинги, векторное хранилище, поисковый конвейер. Первым делом проверяет, нужен ли RAG вашему проекту. CX Codex +1 189установок

Спонсоры проекта

Компании, которые помогают держать аналитику открытой