Overview

This skill streamlines requesting a code review at key points in development. It helps catch issues early and ensure work aligns with requirements before merging or releasing.

How This Skill Works

You define a baseline and end point with git SHAs, then dispatch the code-reviewer subagent using a filled template. The reviewer returns structured feedback, which you triage by priority (Critical, Important, Minor) and address accordingly to keep work compliant with requirements.

When to Use It

• After each task in subagent-driven development
• After completing a major feature
• Before merging to main
• When stuck and you need a fresh perspective
• Before refactoring to establish a baseline

Quick Start

1. Step 1: Determine BASE_SHA (e.g., git rev-parse HEAD~1 or origin/main) and HEAD_SHA (git rev-parse HEAD)
2. Step 2: Dispatch code-reviewer subagent with the Task tool, filling code-reviewer.md using the placeholders
3. Step 3: Act on feedback: fix Critical immediately, fix Important before proceeding, note Minor for later, and push updates (re-run review if needed)

Best Practices

• Capture BASE_SHA and HEAD_SHA for every request
• Fill the code-reviewer.md template precisely with WHAT_WAS_IMPLEMENTED, PLAN_OR_REQUIREMENTS, DESCRIPTION, BASE_SHA, and HEAD_SHA
• Provide a concise DESCRIPTION and clear PLAN_OR_REQUIREMENTS so reviewers can verify scope
• Prioritize feedback: fix Critical first, then Important, and note Minor issues for later
• Push back with reasoning if you disagree with reviewer feedback and document the rationale

Example Use Cases

• After implementing a new authentication feature, request a code review before merging to main
• Before merging a bug fix that touches security-critical code paths
• During a refactor to establish a safe baseline and catch regressions early
• After completing a batch of changes in a multi-task plan to ensure consistency
• When you’re stuck and need a fresh perspective on implementation details

Frequently Asked Questions