code-review

star 1

Review code changes against project standards and ticket requirements. Scores issues by confidence (0-100, reports 75+) and produces a structured report.

bzoran By bzoran schedule Updated 3/5/2026

name: code-review description: Review code changes against project standards and ticket requirements. Scores issues by confidence (0-100, reports 75+) and produces a structured report.

Code Review

Comprehensive code review of current changes against project standards and ticket requirements.

Instructions

When invoked (e.g., /code-review or /code-review PROJ-123):

1. Gather Context

  • Run git diff main...HEAD (or appropriate base branch) to see all changes
  • Read .claude/tickets/{TICKET}/description.md if a ticket number is provided
  • Read CLAUDE.md for project coding standards
  • Identify all files that were modified

2. Review Checklist

For each changed file, evaluate:

Ticket Compliance (if ticket provided)

  • All acceptance criteria are addressed
  • No scope creep - changes match what was requested
  • Edge cases from the ticket are handled

Coding Standards

  • Follows project formatting and style conventions
  • Naming conventions are consistent
  • No unnecessary complexity or over-engineering
  • Error handling is appropriate (not excessive, not missing)

Bugs & Security

  • No logic errors or off-by-one mistakes
  • No null/undefined access without checks
  • No SQL injection, XSS, or command injection vectors
  • No hardcoded secrets or credentials
  • No resource leaks (unclosed connections, streams, etc.)

Architecture

  • Changes follow existing project patterns
  • No unnecessary coupling between modules
  • API contracts maintained (no breaking changes unless intended)

Tests

  • New functionality has corresponding tests
  • Edge cases are covered
  • Tests are meaningful (not just testing that code runs)

3. Score and Report Issues

For each issue found, assign a confidence score (0-100):

  • Only report issues scored 75 or higher - this prevents noise from uncertain findings
  • 90-100: Certain bug, security issue, or standards violation
  • 75-89: Likely issue that should be addressed
  • Below 75: Do not report (too uncertain)

4. Output Format

## Code Review: {TICKET or branch name}

### Summary
{1-2 sentence overview: Approved / Needs Changes / Critical Issues}

### Issues Found
{List by severity, only confidence >= 75}

**Critical** (must fix):
- [{score}] file:line - description
  Recommended fix: ...

**Warning** (should fix):
- [{score}] file:line - description
  Recommended fix: ...

**Suggestion** (nice to have):
- [{score}] file:line - description

### Positive Observations
- What was done well
- Good patterns used
- Improvements over previous approaches

### Files Reviewed
- path/to/file - {brief description of changes}

Important

  • Be thorough but constructive - explain WHY something is an issue
  • Provide concrete code examples for fixes
  • Prioritize by impact - don't let style issues overshadow bugs
  • Acknowledge good code - positive feedback reinforces good practices
  • Only report issues you're confident about (75+ score)
  • Consider the context - a quick fix has different standards than a new feature
Install via CLI
npx skills add https://github.com/bzoran/ai-agent-patterns-and-workflow --skill code-review
Repository Details
star Stars 1
call_split Forks 0
navigation Branch main
article Path SKILL.md
More from Creator