sf-agent-framework/sf-core/tasks/code-review.md

AI Agent Orchestration Framework for Salesforce Development - Two-phase architecture with 70% context reduction

# Code Review

## Purpose

Conduct thorough code reviews of Salesforce implementations to ensure quality,
maintainability, security, and adherence to best practices.

## Instructions

1. **Code Quality Assessment**
   - Review code structure and organization
   - Check naming conventions compliance
   - Assess code readability and documentation
   - Evaluate error handling implementation

2. **Apex Code Review**
   - Verify bulkification of all operations
   - Check governor limit considerations
   - Review SOQL/SOSL query optimization
   - Assess trigger framework usage
   - Validate security enforcement (CRUD/FLS)

3. **Lightning Component Review**
   - Evaluate LWC vs Aura usage decisions
   - Check component composition and reusability
   - Review JavaScript best practices
   - Assess performance optimizations
   - Verify accessibility compliance

4. **Security Review**
   - Check for SOQL injection vulnerabilities
   - Verify sharing and security enforcement
   - Review authentication and authorization
   - Assess data exposure risks
   - Validate input sanitization

5. **Test Coverage Analysis**
   - Verify minimum 75% code coverage
   - Review test quality and assertions
   - Check bulk testing scenarios
   - Assess negative test cases
   - Validate test data creation patterns

6. **Performance Considerations**
   - Review query selectivity
   - Check for inefficient loops
   - Assess batch job design
   - Verify asynchronous processing usage
   - Evaluate caching strategies

## Input Requirements

- Source code repositories
- Code coverage reports
- Static code analysis results
- Architecture design documents
- Coding standards documentation

## Output Format

- Code Review Report containing:
  - Executive summary
  - Detailed findings by category
  - Security vulnerabilities
  - Performance concerns
  - Best practice violations
  - Recommendations with severity
  - Code quality metrics

## Review Checklist

- [ ] Bulkification implemented
- [ ] Governor limits considered
- [ ] Error handling comprehensive
- [ ] Security enforced properly
- [ ] Code well-documented
- [ ] Test coverage adequate
- [ ] No hardcoded values
- [ ] Proper separation of concerns

## Best Practices

- Use automated scanning tools (PMD, ESLint)
- Focus on high-risk areas first
- Provide constructive feedback
- Document patterns for reuse
- Track and measure improvements
- Share knowledge across team