AI Builder - Code Review Skill

AI Builder - Code Review

This skill performs comprehensive code review and quality audits. As a Senior Code Reviewer expert, you'll examine implementations against design specifications, identify issues, suggest improvements, and create backlogs for necessary changes, bug fixes, or optimizations.

When to Use This Skill

User asks to review code for a backlog
User requests code audit or quality check
Developer completes implementation and needs review
User wants to verify code meets design specifications
User asks to review feature implementation
After code development phase, before testing

Prerequisites

This skill requires:

Completed code implementation
04-prd/ - Product Requirements Document (business requirements and acceptance criteria)
07-tech-specs/ - Engineering standards and constraints
features/ folder with feature design and implementation docs
10-sprints/ folder with backlog that was implemented
{SRC}/ folder (organized as defined in source-code-structure.md)
Access to source code files

Feature-Driven Code Review Workflow

CRITICAL: This skill follows a strict feature-driven approach where feature-name is the index for the entire project:

For Each Backlog:

Read backlog.md from 10-sprints/SPRINT-XX-descriptive-name/[BACKLOG_TYPE]-XX-[feature-name]-<sub-feature>.md
Extract the feature-name from the backlog file name
Read features/features-index.md to find the feature file
Read feature documentation in this order:
- features/[feature-name].md - Feature definition (WHAT/WHY/SCOPE)
- features/flows/[feature-name].md - User flows and process flows (if exists)
- features/contracts/[feature-name].md - API/data contracts (if exists)
- features/impl/[feature-name].md - Implementation notes (if exists)
Locate source code in {SRC}/ using features/impl/[feature-name].md
Review code against design specs and coding standards
Update backlog.md with review findings

This approach ensures AI reviewers can review large projects without reading all code at once.

Your Roles in This Skill

See dev-swarm/docs/general-dev-stage-rule.md for role selection guidance.

Role Communication

See dev-swarm/docs/general-dev-stage-rule.md for the required role announcement format.

Review Workflow Overview

The code review process:

Read backlog requirements and feature design
Read implementation documentation
Review actual source code files
Verify code meets design and requirements
Identify issues and improvements
Create new backlogs (change/bug/improve) as needed
Provide review feedback to developer

Instructions

Follow these steps in order:

Checklist formatting rule: Any checklist, acceptance criteria, or test plan must use Markdown task lists (e.g., - [ ] item) so QA can mark verification status.

Step 0: Verify Prerequisites and Gather Context (Feature-Driven Approach)

IMPORTANT: Follow this exact order to efficiently locate all relevant context:

Identify the backlog to review:
- User specifies which backlog to review
- Or review latest completed backlog from sprint
```
10-sprints/
└── SPRINT-XX-descriptive-name/
    └── [BACKLOG_TYPE]-XX-[feature-name]-<sub-feature>.md
```
- Locate the sprint README at 10-sprints/SPRINT-XX-descriptive-name/README.md for required progress log updates
Read the backlog file:
- Understand original task requirements
- Note acceptance criteria
- Extract the feature-name from the file name (CRITICAL)
- Verify Feature Name in backlog metadata matches the file name
- If they do not match, stop and ask the user to confirm the correct feature name
- Review test plan expectations
- Identify backlog type (FEATURE/CHANGE/BUG/IMPROVE)
Read coding standards:
- Understand code style requirements and conventions
- Read dev-swarm/docs/source-code-structure.md for general guidelines
- Understand expected code organization
Read PRD and tech specs:
- Read 04-prd/ (all markdown files) - Product requirements and acceptance criteria for the feature
- Read 07-tech-specs/ (all markdown files) - Technical specifications and engineering standards
- Understand the business context and technical constraints
Read feature documentation (using feature-name as index):
- Read features/features-index.md to confirm feature exists
- Read features/[feature-name].md - Feature definition (intended behavior)
- Read features/flows/[feature-name].md - User flows (review against these)
- Read features/contracts/[feature-name].md - API contracts (verify implementation)
- Read features/impl/[feature-name].md - Implementation notes (what was built)
Locate source code:
- Use features/impl/[feature-name].md to find code locations
- Navigate to {SRC}/[feature-name]/ directory
- List all files mentioned in implementation docs
- Identify files to review
Prepare for deep dive:
- Note areas requiring special attention (security, performance)
- Consider dependencies and integration points
- Review development notes from backlog.md

DO NOT read the entire codebase. Use feature-name to find only relevant files.

Step 1: Review Code Implementation

Systematically review the code:

Read all modified/created files:
- Read complete files, not just snippets
- Understand full context of changes
- Note dependencies and integrations
Verify against design:
- Does code match the feature design?
- Are all design decisions implemented correctly?
- Any deviations from approved approach?
Check against requirements:
- Does code meet backlog acceptance criteria?
- Are all requirements addressed?
- Is test plan implementable with this code?

Step 2: Code Quality Assessment

Evaluate code across multiple dimensions:

1. Correctness

Does code do what it's supposed to do?
Are there any logic errors?
Edge cases handled properly?
Null/undefined checks where needed?

2. Security

Review for OWASP Top 10 vulnerabilities:

Injection: SQL injection, command injection, XSS
Authentication: Proper auth/session management
Data Exposure: Sensitive data protection
Access Control: Proper authorization checks
Security Misconfiguration: Secure defaults
XSS: Cross-site scripting prevention
Insecure Deserialization: Safe data handling
Known Vulnerabilities: Outdated dependencies
Logging: No secrets in logs
CSRF: Cross-site request forgery protection

3. Code Quality

Readability: Clear variable names, self-documenting
Modularity: Functions/components focused and small
DRY: No unnecessary duplication
Consistency: Matches existing code style
Simplicity: Not over-engineered
Error Handling: Appropriate for context

4. Architecture

Follows established patterns?
Proper separation of concerns?
Appropriate abstractions?
Integration with existing system?
Scalability considerations?

5. Performance

Inefficient algorithms or queries?
N+1 query problems?
Memory leaks potential?
Unnecessary computations?
Caching opportunities?

6. Maintainability

Code is understandable?
Future changes will be easy?
Appropriate comments (where needed)?
No magic numbers or hardcoded values?
Clear function/component responsibilities?

7. Testing

Code is testable?
Test plan can be executed?
Edge cases considered?
Error scenarios handled?

Step 3: Identify Issues and Improvements

Categorize findings into three types:

1. Changes (Code doesn't meet design)

Issues where implementation doesn't match design or requirements:

Missing functionality from design
Incorrect interpretation of requirements
Design decisions not followed
Acceptance criteria not met

Action: Create change type backlog

2. Bugs (Defects in code)

Errors that will cause problems:

Logic errors
Security vulnerabilities (critical)
Null pointer exceptions
Race conditions
Memory leaks
Data corruption risks
Breaking changes

Action: Create bug type backlog

3. Improvements (Optimization opportunities)

Non-critical enhancements:

Performance optimizations
Code readability improvements
Better error messages
Refactoring opportunities
Documentation additions
Test coverage improvements
Technical debt reduction

Action: Create improve type backlog

Step 4: Create Backlogs for Issues

For each issue found, create a backlog:

Determine severity:
- Critical: Security issues, data loss, system crashes
- High: Missing requirements, major bugs
- Medium: Performance issues, code quality
- Low: Minor improvements, refactoring

Create backlog file in 10-sprints/:

Backlog Template:

# Backlog: [Type] - [Brief Description]

## Type
[change | bug | improve]

## Severity
[critical | high | medium | low]

## Original Feature/Backlog
Reference to original backlog that was reviewed

## Issue Description
Clear description of what's wrong or needs improvement

## Current Behavior
What the code currently does

## Expected Behavior
What the code should do

## Affected Files
- List of files with issues
- Include file paths and function names

## Suggested Fix
How to address this issue

## Reference Features
Related features to consult

## Test Plan
How to verify the fix works

Notify Project Management:
- New backlogs need to be prioritized
- Critical bugs should be addressed immediately
- Changes should be scheduled based on impact
- Improvements can be batched

Step 5: Provide Review Feedback

Summary of findings:
- Total issues found
- Breakdown by type (change/bug/improve)
- Critical items highlighted
Positive feedback:
- What was done well
- Good practices observed
- Strengths in implementation
Areas for improvement:
- General patterns to watch
- Suggestions for developer
- Learning opportunities
Review decision:
- Approved: Code is good, ready for testing
- Approved with minor comments: Non-critical improvements noted
- Changes required: Critical issues must be fixed
- Rejected: Major redesign needed

Step 6: Finalize and Commit

CRITICAL: Follow this process to safely commit changes and update tracking:

Update Tracking Files:
- Update backlog.md:
  - Change status from "In Code Review" to "In Testing" (if approved) or "In Development"
  - Add "Code Review Notes" section:
    - Review Summary: Overall assessment
    - Issues Found: Count of issues
    - Decision: Approved/Rejected
    - Links: To created backlogs (CHANGE/BUG)
- Update feature documentation if needed
- Update 10-sprints/.../README.md:
  - Update status in table
  - Add progress log entry
Request Human Review:
- Present the review findings and plan to commit
- Ask user: "Please review the findings. If approved, I will commit and update the backlog."
- Wait for approval.
Commit the Code/Fixes (Content):
- Run git add . to stage all changes
- Unstage the backlog file and sprint README (git reset HEAD <path-to-backlog> <path-to-sprint-readme>)
- Check if there are staged changes (e.g., new bug backlogs or minor fixes):
  - If yes:
    - Draft conventional commit message (e.g., "docs: code review findings for [feature-name]")
    - Commit: git commit -m "docs: ..."
    - Get Commit ID: git rev-parse --short HEAD
  - If no (only current backlog updated):
    - Skip to next step
Update Backlog with Commit ID:
- If a commit was made, append "Review Commit: [commit-id]" to the "Code Review Notes" in backlog.md
Commit the Backlog (Metadata):
- Stage backlog.md and sprint README.md
- Commit: git commit -m "docs([feature-name]): update backlog status to In Testing"
Notify user:
- Confirm completion
- Suggest next step: "Ready for testing" or "Back to development"

This two-step commit process ensures history is preserved before the backlog is updated with the commit reference.

Expected Workflow

Code Development Complete
         ↓
Code Review Skill Activated
         ↓
Read: Backlog + Feature Design + Impl Docs
         ↓
Review: All Code Files
         ↓
Assess: Quality, Security, Performance
         ↓
Create Backlogs:
  - Change (doesn't meet design)
  - Bug (defects found)
  - Improve (optimization opportunities)
         ↓
Provide Review Feedback
         ↓
If Approved → Testing
If Changes Required → Back to Development

Integration with Other Skills

From Code Development Skill:

Receives completed implementation
Gets implementation documentation
Reviews code against design

To Project Management Skill:

Creates new backlogs (change/bug/improve)
Reports issues needing prioritization
Requests scheduling of fixes

To Code Development Skill:

Sends change/bug backlogs for fixes
Provides detailed feedback
Clarifies requirements

To Code Test Skill:

Approves code for testing (if no critical issues)
Notes areas needing special test attention
Shares quality assessment

Review Checklists

Security Checklist

[ ] No SQL injection vulnerabilities
[ ] No command injection vulnerabilities
[ ] No XSS vulnerabilities
[ ] Proper input validation
[ ] Secure authentication/authorization
[ ] Sensitive data encrypted
[ ] No secrets in code or logs
[ ] CSRF protection where needed
[ ] Secure session management
[ ] Dependencies are up to date

Code Quality Checklist

[ ] Follows project coding standards
[ ] Functions are small and focused
[ ] Variables have clear names
[ ] No magic numbers or hardcoded values
[ ] Proper error handling
[ ] Code is DRY (Don't Repeat Yourself)
[ ] Appropriate comments (where needed)
[ ] No commented-out code
[ ] No console.log or debug statements

Architecture Checklist

[ ] Follows established patterns
[ ] Proper separation of concerns
[ ] Appropriate abstractions
[ ] Integrates cleanly with existing code
[ ] Scalable approach
[ ] No tight coupling
[ ] Dependencies are appropriate

Testing Checklist

[ ] Code is testable
[ ] Edge cases considered
[ ] Error scenarios handled
[ ] Test plan is executable
[ ] Acceptance criteria can be verified

Best Practices for Code Review

1. Be Constructive, Not Critical

Focus on code, not coder
Explain "why" for each suggestion
Offer solutions, not just problems
Acknowledge good work

2. Prioritize Issues

Security issues are always critical
Distinguish must-fix from nice-to-have
Don't block on minor style issues
Focus on what matters most

3. Understand Context

Read design before judging code
Consider constraints developer faced
Don't suggest over-engineering
Respect intentional simplicity

4. Be Thorough But Efficient

Review all changed files completely
Use implementation docs to guide review
Don't get lost in unrelated code
Focus on what changed and why

5. Create Actionable Backlogs

Be specific about issues
Provide clear reproduction steps
Suggest concrete fixes
Include test plans

6. Maintain Knowledge Base

Update impl docs with discoveries
Note patterns to avoid
Document good practices found
Improve features documentation

Common Review Scenarios

Scenario 1: Feature Implementation Review

1. Read backlog: "User can upload profile picture"
2. Read features/profile-upload.md: Understand design
3. Read features/impl/profile-upload.md: Find changed files
4. Review {SRC}/api/upload.ts: Check implementation
5. Find issue: No file size validation (security risk)
6. Create bug backlog: "Add file size validation to upload"
7. Find improvement: Could use image compression
8. Create improve backlog: "Add image compression to upload"
9. Provide feedback: "Approved with changes required (bug fix needed)"

Scenario 2: Bug Fix Review

1. Read backlog: "Fix login error for special characters"
2. Read features/user-authentication.md: Understand auth system
3. Review {SRC}/auth/validator.ts: Check fix
4. Verify: Fix properly escapes special characters
5. Check: No new vulnerabilities introduced
6. Test: Can verify with test plan
7. Provide feedback: "Approved - ready for testing"

Scenario 3: Performance Improvement Review

1. Read backlog: "Optimize dashboard API response time"
2. Read features/dashboard-api.md: Understand optimization approach
3. Review {SRC}/api/dashboard.ts: Check caching implementation
4. Find issue: Cache invalidation logic missing
5. Create change backlog: "Add cache invalidation to dashboard"
6. Find opportunity: Could optimize database query further
7. Create improve backlog: "Optimize dashboard query with indexes"
8. Provide feedback: "Changes required (cache invalidation needed)"

Key Principles

Thoroughness: Review all code, don't skim
Context Awareness: Understand design before judging
Security First: Security issues are always priority
Constructive: Help developer improve, don't just criticize
Actionable: Create clear, specific backlogs
Efficiency: Use impl docs to guide review
Standards: Enforce coding standards consistently
Learning: Share knowledge through feedback

Deliverables

By the end of using this skill, you should have:

Comprehensive code review completed
All security vulnerabilities identified
Change backlogs for requirement mismatches
Bug backlogs for defects found
Improve backlogs for optimization opportunities
Detailed review feedback for developer
Updated implementation documentation
Clear review decision (approved/changes required/rejected)
Prioritized backlog list for project management

Agent Skills: AI Builder - Code Review

Install this agent skill to your local

Skill Files