review-commits

Coordinate parallel AI code reviews against a task's git diff.

Updated Mar 18, 2026
One-click install
npx skills add https://github.com/gioe/python-libs --skill review-commits
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: review-commits
Source: https://github.com/gioe/python-libs/tree/main/.claude/skills/review-commits
Command: npx skills add https://github.com/gioe/python-libs --skill review-commits

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill automates the orchestration of code reviews by running parallel AI reviewer agents against a task's git diff, surfacing must_fix issues and coordinating defer/suggest outcomes.

Core Features & Use Cases

  • Spawns one background reviewer agent per enabled reviewer from config.
  • Analyzes diffs between the current branch and base to identify correctness, security, and maintainability issues.
  • Aggregates reviewer findings, applies must_fix fixes, and handles suggest/defer flows interactively.
  • Creates deferred tasks for defer findings to enable later work while keeping the main PR moving.

Quick Start

Run the /review-commits command against the current branch to start parallel AI code reviews against the diff.

Frequently Asked Questions about review-commits

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

FAQPage Schema
How do I automate AI code reviews on my git diff?

Automating AI code reviews on a git diff involves spawning parallel reviewer agents per enabled configuration. This Skill orchestrates agent reviews against the base-branch diff range, collecting findings and applying must_fix fixes automatically.

Can I run parallel AI reviewers against a specific base branch range?

Yes, running parallel AI reviewers against a base branch range analyzes diffs for correctness and security issues. The process compares the current branch against the base to identify problems and aggregates the findings.

What is the best way to handle non-blocking issues found during an AI code review?

Handling non-blocking issues found during an AI code review uses defer or suggest decisions to keep the main PR moving. The mechanism creates deferred tasks for defer findings, enabling later work without blocking the current branch.

How do I apply must_fix corrections identified by automated git diff reviewers?

Applying must_fix corrections identified by automated git diff reviewers is handled by aggregating findings and applying fixes directly. The system coordinates the interactive suggest and defer flows while applying necessary changes.

Does this parallel review approach require manual configuration for each reviewer agent?

This parallel review approach does not require manual configuration for each reviewer agent, as it spawns one background reviewer per enabled reviewer from config. It automatically orchestrates the review process against the task's git diff.