What should I do if I disagree with the AI reviewer?

The skill encourages active collaboration; if the reviewer is wrong, you should push back with technical reasoning or provide updated tests to clarify the implementation.

Is this skill specific to any programming language?

While optimized for the 'indie-stack' (Swift, Supabase, Vercel), the core code-review logic can be applied to any codebase managed within Claude Code.

When is the best time to trigger a review?

It is best used after completing a specific task, before merging a feature branch, or whenever you need a fresh perspective on a complex implementation.

How does the subagent know which code to review?

The skill utilizes git SHAs to pinpoint the exact range of changes between your starting point (Base SHA) and current work (Head SHA).

Automated Code Reviewer

Name: Automated Code Reviewer
Author: bgrober

bybgrober

•

Collaboration Tools

Automates the code review process by dispatching specialized subagents to verify implementation requirements and catch issues before merging.

About

This skill integrates a sophisticated 3-stage review process into the Claude Code workflow, designed specifically for indie developers. It allows users to dispatch a 'superpowers:code-reviewer' subagent that analyzes code changes against specific plans and requirements using git SHAs. By categorizing feedback into Critical, Important, and Minor tiers, it ensures that technical debt and UX issues are addressed early in the development cycle. Whether you are following a subagent-driven development pattern or working ad-hoc, this skill provides a structured framework for maintaining high code quality and architectural consistency.

Key Features

1 GitHub stars
Automated subagent dispatch for deep code analysis
Template-driven reviews based on project requirements
Tiered feedback system (Critical, Important, and Minor issues)
Git-integrated change isolation using Base and Head SHAs
UX-focused evaluation for frontend and mobile components

Use Cases

Performing final quality assurance checks before merging to the main branch
Verifying task completion in subagent-driven development cycles
Getting automated technical feedback when stuck on complex debugging tasks

About

Key Features

1 GitHub stars
Automated subagent dispatch for deep code analysis
Template-driven reviews based on project requirements
Tiered feedback system (Critical, Important, and Minor issues)
Git-integrated change isolation using Base and Head SHAs
UX-focused evaluation for frontend and mobile components

Use Cases

Performing final quality assurance checks before merging to the main branch
Verifying task completion in subagent-driven development cycles
Getting automated technical feedback when stuck on complex debugging tasks