Files
gh-jamesrochabrun-skills-al…/skills/anthropic-architect/references/use_case_examples.md
2025-11-29 18:48:58 +08:00

1001 lines
20 KiB
Markdown

# Real-World Use Case Examples
Comprehensive examples of Anthropic architectures for common use cases, with detailed implementation guidance.
## Table of Contents
1. [Code Quality & Review](#code-quality--review)
2. [Feature Development](#feature-development)
3. [Documentation](#documentation)
4. [Testing & QA](#testing--qa)
5. [Refactoring & Modernization](#refactoring--modernization)
6. [Security & Compliance](#security--compliance)
7. [DevOps & Automation](#devops--automation)
8. [Data Analysis](#data-analysis)
9. [Customer Support](#customer-support)
10. [Content Generation](#content-generation)
---
## Code Quality & Review
### Use Case 1: Automated Code Review System
**Requirements:**
- Review all PRs automatically
- Check coding standards
- Identify security issues
- Suggest improvements
- Team-wide consistency
**Complexity Analysis:**
- Task Complexity: 6 (structured workflow)
- Reusability: 8 (organization-wide)
- Context: 5 (needs coding standards)
- Security: 5 (read operations)
**Recommended Architecture:** **Skills + Direct Prompts**
**Implementation:**
**1. Create code-review-expert skill:**
```
code-review-expert/
├── SKILL.md
├── references/
│ ├── coding_standards.md
│ ├── security_patterns.md
│ ├── performance_guidelines.md
│ └── review_checklist.md
└── scripts/
└── review_pr.sh
```
**2. Usage for PRs:**
```
Simple PR (< 50 lines):
→ Direct Prompt
"Using code-review-expert skill, review PR #123"
Complex PR (> 500 lines):
→ Agent + Skill
Review Agent:
- Load code-review-expert skill
- Analyze all changed files
- Generate comprehensive review
```
**3. coding_standards.md:**
```markdown
# Team Coding Standards
## TypeScript
- Use strict mode
- Explicit return types
- No `any` types
- Prefer const over let
- 100 character line limit
## Patterns
- Repository pattern for data access
- Dependency injection for services
- Factory pattern for object creation
## Security
- Validate all inputs
- Sanitize user data
- Use parameterized queries
- Never log sensitive data
```
**4. Review Workflow:**
```bash
#!/bin/bash
# review_pr.sh
PR_NUMBER=$1
echo "Reviewing PR #$PR_NUMBER..."
# Get PR diff
gh pr diff $PR_NUMBER > /tmp/pr_diff.txt
# Review with Claude + skill
claude "Using code-review-expert skill, review this PR:
Changes:
$(cat /tmp/pr_diff.txt)
Check for:
1. Coding standards compliance
2. Security vulnerabilities
3. Performance issues
4. Best practices adherence
5. Test coverage
Format: Provide detailed review with specific line numbers and suggestions."
```
**Benefits:**
- Consistent reviews
- Team standards enforced
- Reusable across all PRs
- Fast feedback
**Metrics:**
- Review time: 2-5 minutes
- Coverage: 100% of PRs
- False positives: < 5%
- Developer satisfaction: High
---
### Use Case 2: Code Quality Dashboard
**Requirements:**
- Analyze entire codebase
- Track metrics over time
- Identify problem areas
- Generate reports
**Recommended Architecture:** **Parallel Agents**
**Implementation:**
```
Quality Dashboard Workflow
Coordinator Agent
├─┬─ Complexity Agent → Analyze code complexity
│ │
│ ├─ Coverage Agent → Check test coverage
│ │
│ ├─ Security Agent → Find vulnerabilities
│ │
│ └─ Duplication Agent → Detect code duplication
└─→ Dashboard Agent → Generate visual report
All agents run in parallel → 4x faster
```
**Code:**
```typescript
// Launch parallel agents
const [
complexityReport,
coverageReport,
securityReport,
duplicationReport
] = await Promise.all([
analyzeComplexity(codebase),
analyzeCoverage(codebase),
analyzeSecurity(codebase),
analyzeDuplication(codebase)
]);
// Aggregate results
const dashboard = generateDashboard({
complexity: complexityReport,
coverage: coverageReport,
security: securityReport,
duplication: duplicationReport
});
```
---
## Feature Development
### Use Case 3: Full-Stack Feature Implementation
**Requirements:**
- Implement user authentication
- Frontend + Backend + Database
- Tests and documentation
- Deploy to staging
**Recommended Architecture:** **Specialist Agent Team + Skills**
**Implementation:**
```
Project Lead Agent
├── Frontend Agent
│ Load: frontend-designer-skill
│ Tasks:
│ - Create login/signup components
│ - Add form validation
│ - Implement token storage
│ - Add loading states
├── Backend Agent
│ Load: api-design-expert-skill
│ Tasks:
│ - Create auth endpoints
│ - Implement JWT generation
│ - Add refresh token logic
│ - Database queries
├── Database Agent
│ Load: database-expert-skill
│ Tasks:
│ - Design user schema
│ - Create migrations
│ - Add indexes
│ - Seed test data
├── Testing Agent
│ Load: testing-strategies-skill
│ Tasks:
│ - Unit tests
│ - Integration tests
│ - E2E tests
│ - Coverage reports
└── DevOps Agent
Load: deployment-expert-skill
Tasks:
- Update configs
- Deploy to staging
- Run smoke tests
- Monitor deployment
```
**Execution:**
```
Phase 1: Planning (Sequential)
Lead Agent creates detailed plan
Phase 2: Implementation (Parallel)
All specialist agents work simultaneously
- Frontend: 15 minutes
- Backend: 15 minutes
- Database: 10 minutes
- Tests: 20 minutes
Total: 20 minutes (vs 70 minutes sequential)
Phase 3: Integration (Sequential)
Lead Agent coordinates integration
Phase 4: Deployment (Sequential)
DevOps Agent deploys to staging
```
**Benefits:**
- Parallel execution (70% faster)
- Specialized expertise per layer
- Comprehensive test coverage
- Automated deployment
---
### Use Case 4: Feature Flag System
**Requirements:**
- Add feature flag capability
- Minimal risk implementation
- Gradual rollout support
- Analytics integration
**Recommended Architecture:** **Skills + Multi-Phase Agents**
**Implementation:**
```
Phase 1: Exploration Agent
Tools: [Read, Grep, Glob]
Load: architecture-patterns-skill
Task:
- Analyze current architecture
- Identify integration points
- Research feature flag patterns
- Assess risk areas
Phase 2: Design Agent
Tools: [TodoWrite]
Load: system-design-skill
Task:
- Design feature flag service
- Plan database schema
- Define API contracts
- Create rollout strategy
Phase 3: Implementation Agent
Tools: [Read, Edit, Write]
Load: coding-standards-skill
Task:
- Implement feature flag service
- Add database migrations
- Create admin UI
- Write documentation
Phase 4: Testing Agent
Tools: [Bash, Read]
Load: testing-strategies-skill
Task:
- Unit tests
- Integration tests
- Load testing
- Rollback testing
Phase 5: Deployment Agent
Tools: [Bash]
Load: deployment-patterns-skill
Confirmations: ['deployment', 'database migration']
Task:
- Deploy to staging
- Run validation tests
- Deploy to production (10% users)
- Monitor metrics
```
---
## Documentation
### Use Case 5: API Documentation Generator
**Requirements:**
- Generate API docs from code
- Include examples
- Keep docs in sync with code
- Multiple output formats
**Recommended Architecture:** **Skills**
**Implementation:**
```
api-documentation-skill/
├── SKILL.md
├── references/
│ ├── openapi_spec.md
│ ├── documentation_guidelines.md
│ └── example_formats.md
├── templates/
│ ├── endpoint_template.md
│ ├── schema_template.md
│ └── example_template.md
└── scripts/
└── generate_docs.sh
```
**Usage:**
```
Using api-documentation-skill, generate complete API
documentation for all endpoints in src/api/
Include:
- OpenAPI spec
- Request/response examples
- Authentication details
- Error responses
- Rate limiting info
Format: Markdown with code examples
```
**Output:**
```markdown
# User API
## POST /api/users
Create a new user account.
### Request
```json
{
"email": "user@example.com",
"password": "SecurePass123!",
"name": "John Doe"
}
```
### Response (201 Created)
```json
{
"id": "usr_123",
"email": "user@example.com",
"name": "John Doe",
"createdAt": "2025-01-15T10:30:00Z"
}
```
### Errors
- 400: Invalid email format
- 409: Email already exists
- 429: Rate limit exceeded
```
---
### Use Case 6: Onboarding Documentation
**Requirements:**
- Help new developers get started
- Multiple learning paths
- Interactive tutorials
- Up-to-date with codebase
**Recommended Architecture:** **Progressive Disclosure Skill**
**Implementation:**
```
onboarding-guide/
├── SKILL.md
├── index.md (learning paths)
├── getting_started/
│ ├── day_1.md
│ ├── day_2.md
│ └── week_1.md
├── by_role/
│ ├── frontend_dev.md
│ ├── backend_dev.md
│ └── fullstack_dev.md
├── by_topic/
│ ├── architecture.md
│ ├── deployment.md
│ └── testing.md
└── interactive/
├── setup_walkthrough.sh
└── first_pr_guide.sh
```
**Progressive Disclosure:**
```
New developer: "I'm a frontend developer, where do I start?"
1. Load: index.md
→ Shows: Learning paths
2. Load: by_role/frontend_dev.md
→ Shows: Frontend-specific onboarding
3. Load: getting_started/day_1.md
→ Shows: Day 1 tasks
4. As they progress:
→ Load additional topics on demand
```
---
## Testing & QA
### Use Case 7: Comprehensive Test Suite Generator
**Requirements:**
- Generate tests for new features
- Multiple test types
- High coverage targets
- Integration with CI/CD
**Recommended Architecture:** **Agent + Skills**
**Implementation:**
```
Test Generation Agent
Load Skills:
├── testing-strategies-skill
└── code-analysis-skill
Workflow:
1. Analyze code to test
2. Identify test scenarios
3. Generate unit tests
4. Generate integration tests
5. Generate E2E tests
6. Verify coverage > 80%
```
**Example:**
```
Using testing-strategies-skill, generate comprehensive test suite
for UserService class:
Coverage requirements:
- Unit tests: 90%+
- Integration tests: All API endpoints
- E2E tests: Critical user flows
Test frameworks:
- Jest for unit/integration
- Playwright for E2E
Include:
- Happy paths
- Error cases
- Edge cases
- Security tests
```
**Generated Output:**
```typescript
// user-service.test.ts
describe('UserService', () => {
describe('createUser', () => {
it('should create user with valid data', async () => {
// Test implementation
});
it('should reject invalid email', async () => {
// Test implementation
});
it('should reject duplicate email', async () => {
// Test implementation
});
it('should hash password', async () => {
// Test implementation
});
});
// ... more tests
});
```
---
### Use Case 8: Regression Test Suite Maintenance
**Requirements:**
- Keep tests up-to-date
- Remove obsolete tests
- Update for API changes
- Improve flaky tests
**Recommended Architecture:** **Multi-Phase Agents**
**Implementation:**
```
Phase 1: Analysis Agent
- Identify failing tests
- Find flaky tests
- Detect obsolete tests
- Check coverage gaps
Phase 2: Cleanup Agent
- Remove obsolete tests
- Update outdated assertions
- Fix broken imports
Phase 3: Enhancement Agent
- Improve flaky tests
- Add missing test cases
- Increase coverage
Phase 4: Validation Agent
- Run full test suite
- Verify improvements
- Generate report
```
---
## Refactoring & Modernization
### Use Case 9: Legacy Code Modernization
**Requirements:**
- Upgrade from JavaScript to TypeScript
- Update deprecated APIs
- Improve code quality
- Maintain functionality
**Recommended Architecture:** **Multi-Phase Agents + Skills**
**Implementation:**
```
Modernization Pipeline
Phase 1: Assessment Agent
Load: code-quality-skill
- Analyze current codebase
- Identify modernization opportunities
- Assess risk levels
- Create priority list
Phase 2: Planning Agent
Load: refactoring-patterns-skill
- Create detailed migration plan
- Define phases and milestones
- Identify dependencies
- Plan rollback strategy
Phase 3: TypeScript Migration Agent
Load: typescript-migration-skill
- Convert .js to .ts
- Add type definitions
- Fix type errors
- Update build config
Phase 4: API Update Agent
Load: api-modernization-skill
- Replace deprecated APIs
- Update to new patterns
- Modernize syntax
- Optimize performance
Phase 5: Testing Agent
Load: testing-strategies-skill
- Run existing tests
- Add new tests
- Verify functionality
- Check performance
Phase 6: Validation Agent
- Compare behavior (before/after)
- Run regression tests
- Performance benchmarks
- Generate migration report
```
**Safety Measures:**
```
- Git branch per phase
- Tests after each phase
- Rollback capability
- Progressive rollout
- Monitoring and alerts
```
---
### Use Case 10: Microservices Extraction
**Requirements:**
- Extract service from monolith
- Minimal downtime
- Data migration
- Gradual rollout
**Recommended Architecture:** **SDK Custom Workflow**
**Implementation:**
```typescript
const microserviceExtraction = {
phases: [
{
name: 'Analysis',
agent: analysisAgent,
tasks: [
'Identify service boundaries',
'Map dependencies',
'Plan data separation',
'Design API contracts'
]
},
{
name: 'Preparation',
agent: prepAgent,
tasks: [
'Create new service structure',
'Set up database',
'Implement API',
'Add monitoring'
]
},
{
name: 'Migration',
agent: migrationAgent,
confirmations: ['database migration', 'traffic routing'],
tasks: [
'Migrate data',
'Deploy service',
'Route 10% traffic',
'Monitor metrics'
]
},
{
name: 'Validation',
agent: validationAgent,
tasks: [
'Compare responses',
'Check performance',
'Verify data consistency',
'Gradual increase to 100%'
]
}
]
};
```
---
## Security & Compliance
### Use Case 11: Security Audit System
**Requirements:**
- Automated security scans
- OWASP compliance
- Vulnerability detection
- Remediation guidance
**Recommended Architecture:** **Specialist Agents + Security Skill**
**Implementation:**
```
Security Audit Coordinator
├── OWASP Scanner Agent
│ Load: owasp-security-skill
│ Check: OWASP Top 10 vulnerabilities
├── Dependency Scanner Agent
│ Load: dependency-security-skill
│ Check: Vulnerable dependencies
├── Code Scanner Agent
│ Load: secure-coding-skill
│ Check: Code vulnerabilities
├── Config Scanner Agent
│ Load: secure-config-skill
│ Check: Insecure configurations
└── Report Agent
Aggregate: All findings
Prioritize: By severity
Recommend: Remediation steps
```
**Security Skill Structure:**
```
owasp-security-skill/
├── SKILL.md
├── references/
│ ├── owasp_top_10_2024.md
│ ├── vulnerability_patterns.md
│ ├── remediation_guides.md
│ └── secure_coding_practices.md
└── checklist/
├── injection.md
├── authentication.md
├── sensitive_data.md
└── ...
```
---
## DevOps & Automation
### Use Case 12: CI/CD Pipeline Generator
**Requirements:**
- Generate CI/CD pipelines
- Multiple platforms (GitHub, GitLab, etc.)
- Best practices built-in
- Customizable workflows
**Recommended Architecture:** **Skills**
**Implementation:**
```
cicd-generator-skill/
├── SKILL.md
├── references/
│ ├── pipeline_patterns.md
│ ├── best_practices.md
│ └── optimization_tips.md
├── templates/
│ ├── github_actions/
│ │ ├── basic.yml
│ │ ├── advanced.yml
│ │ └── monorepo.yml
│ ├── gitlab_ci/
│ └── jenkins/
└── scripts/
└── generate_pipeline.sh
```
**Usage:**
```
Using cicd-generator-skill, generate GitHub Actions pipeline for:
Project type: Node.js API
Requirements:
- Lint on PR
- Run tests (unit, integration)
- Build Docker image
- Deploy to staging (on main branch)
- Deploy to production (on release tag)
- Security scanning
- Code coverage reports
Include:
- Caching for faster builds
- Parallel jobs where possible
- Proper secrets management
```
**Generated Pipeline:**
```yaml
name: CI/CD Pipeline
on:
pull_request:
branches: [main]
push:
branches: [main]
tags: ['v*']
jobs:
lint:
runs-on: ubuntu-latest
steps:
- uses: actions/checkout@v3
- uses: actions/setup-node@v3
with:
node-version: '18'
cache: 'npm'
- run: npm ci
- run: npm run lint
test:
runs-on: ubuntu-latest
steps:
- uses: actions/checkout@v3
- uses: actions/setup-node@v3
with:
node-version: '18'
cache: 'npm'
- run: npm ci
- run: npm test -- --coverage
- uses: codecov/codecov-action@v3
security:
runs-on: ubuntu-latest
steps:
- uses: actions/checkout@v3
- run: npm audit
- uses: snyk/actions/node@master
# ... more jobs
```
---
## Data Analysis
### Use Case 13: Log Analysis and Insights
**Requirements:**
- Analyze application logs
- Identify patterns
- Detect anomalies
- Generate insights
**Recommended Architecture:** **Agents + Progressive Disclosure Skill**
**Implementation:**
```
Log Analysis Pipeline
1. Collection Agent
- Gather logs from various sources
- Normalize formats
- Filter relevant entries
2. Pattern Recognition Agent
Load: log-analysis-patterns-skill
- Identify error patterns
- Find performance bottlenecks
- Detect security threats
- Track user behavior
3. Anomaly Detection Agent
- Compare against baselines
- Flag unusual patterns
- Assess severity
4. Insights Agent
- Aggregate findings
- Generate recommendations
- Create visualizations
- Prioritize actions
```
**Log Analysis Skill:**
```
log-analysis-patterns-skill/
├── patterns/
│ ├── error_patterns.md
│ ├── performance_patterns.md
│ ├── security_patterns.md
│ └── user_behavior_patterns.md
└── insights/
├── common_issues.md
└── remediation_steps.md
```
---
## Content Generation
### Use Case 14: Technical Blog Post Generator
**Requirements:**
- Generate blog posts from features
- SEO optimized
- Brand voice consistency
- Code examples included
**Recommended Architecture:** **Skills**
**Implementation:**
```
technical-blog-generator-skill/
├── SKILL.md
├── references/
│ ├── brand_voice.md
│ ├── seo_guidelines.md
│ ├── blog_structure.md
│ └── code_example_formats.md
├── templates/
│ ├── feature_announcement.md
│ ├── tutorial.md
│ ├── deep_dive.md
│ └── comparison.md
└── examples/
└── published_posts.md
```
**Usage:**
```
Using technical-blog-generator-skill, create a blog post about
our new API rate limiting feature:
Target audience: Backend developers
Tone: Technical but accessible
Include:
- Why rate limiting matters
- How to implement
- Code examples (Node.js, Python)
- Best practices
- Common pitfalls
SEO keywords: API rate limiting, request throttling, API security
```
---
## Summary
**Key Patterns by Use Case:**
| Use Case | Architecture | Why |
|----------|-------------|-----|
| Code Review | Skills + Prompts | Reusable standards, simple workflow |
| Feature Development | Specialist Agents + Skills | Complex, multi-layer, parallel work |
| Documentation | Progressive Disclosure Skill | Large knowledge base, context efficiency |
| Testing | Agents + Skills | Autonomous generation, quality standards |
| Refactoring | Multi-Phase Agents | Complex, risky, needs validation |
| Security | Specialist Agents + Skills | Multiple scan types, expertise needed |
| CI/CD | Skills | Template generation, best practices |
| Data Analysis | Agents + Skills | Pattern recognition, insights |
| Content | Skills | Brand consistency, SEO requirements |
**Decision Framework:**
1. Analyze your requirements
2. Match to similar use case
3. Adapt architecture pattern
4. Implement and iterate
5. Measure and optimize
**Remember:** These are starting points. Adapt based on your specific needs and constraints.