review-code

User Input

$ARGUMENTS

You MUST consider the user input before proceeding (if not empty).

Outline

1. Initialize Review Context:

   • Run .specify/scripts/bash/check-prerequisites.sh --json --require-tasks --include-tasks from repo root
   • Parse FEATURE_DIR and AVAILABLE_DOCS from output
   • Identify the feature scope and related artifacts

2. Collect Git Changes:

   • Use get_changed_files to retrieve all staged and unstaged changes
   • Parse the diff output to identify:
     • New files created
     • Modified files
     • Deleted files
     • Lines added/removed per file
   • Categorize changes by type: source code, tests, documentation, configuration

3. Load Reference Artifacts:

   • REQUIRED: Read tasks.md to understand what was supposed to be implemented
   • REQUIRED: Read spec.md for feature requirements and acceptance criteria
   • IF EXISTS: Read plan.md for architecture and design decisions
   • IF EXISTS: Read contracts/ for API specifications
   • IF EXISTS: Read data-model.md for entity definitions
   • IF EXISTS: Read checklists/ for quality checkpoints

4. Code Quality Review:

   • Style and Conventions:

     • Check for consistent naming conventions (camelCase, PascalCase, snake_case)
     • Verify proper indentation and formatting
     • Identify unused imports or variables
     • Check for hardcoded values that should be constants/config

   • Code Structure:

     • Verify proper separation of concerns
     • Check for appropriate error handling (try-catch, error types)
     • Identify overly complex functions (>50 lines, high cyclomatic complexity)
     • Verify proper use of async/await patterns
     • Check for proper resource cleanup (connections, streams, timers)

   • Security and Best Practices:

     • Check for exposed secrets or credentials in code
     • Verify input validation and sanitization
     • Check for SQL injection or XSS vulnerabilities
     • Verify proper authentication/authorization checks
     • Check for insecure dependencies or known CVEs