dev-review

Detect changed files, run parallel reviewers, and synthesize prioritized findings.

Updated Aug 23, 2026
One-click install
npx skills add https://github.com/mattjefferson/agents --skill dev-review-mattjefferson
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: dev-review
Source: https://github.com/mattjefferson/agents/tree/main/skills/dev-review
Command: npx skills add https://github.com/mattjefferson/agents --skill dev-review-mattjefferson

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Discover applicable reviewers from a codebase's tech stack and changes, run them as parallel subagents, and synthesize findings into a prioritized list to guide fixes after development work.

Core Features & Use Cases

  • Scope detection: compute diff, extract changed files, and identify touched directories to scope the review.
  • Parallel reviewer orchestration: spawn built-in reviewers and conditional delegated skills to run concurrently.
  • Synthesis & presentation: deduplicate findings, group by severity, and present actionable recommendations and next steps.

Quick Start

Initiate a dev-review on the current branch to obtain a prioritized list of findings and recommended fixes.

Frequently Asked Questions about dev-review

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I automate code reviews to run multiple checks in parallel?

You can automate code reviews by orchestrating subagents that identify applicable reviewers from your diff and run them concurrently. This parallel orchestration synthesizes findings into a prioritized list, deduplicating results to guide fixes.

What is the best way to scope a code review to only changed files?

Scoping a code review to changed files involves computing the diff, extracting touched files, and identifying affected directories. This scope detection ensures reviewers only analyze relevant code changes.

How does code review synthesis group findings by severity?

Code review synthesis groups findings by severity by deduplicating detected issues and organizing them into a structured report. This process presents actionable recommendations and a final verdict with next steps.

Can I use parallel subagents to review code changes without manual reviewer selection?

Yes, you can use parallel subagents to review code changes without manual selection. The system detects your tech stack and changes to automatically select built-in reviewers and delegated skills.

Do I need to configure specific reviewers before running a code review workflow?

No, you do not need to configure specific reviewers beforehand. The workflow automatically discovers applicable reviewers from the codebase's tech stack and current changes to run them concurrently.