Review Pr

This is a structured PR review system that deploys multiple specialized agents to analyze your code from different angles: security, bugs, code quality, test coverage, API contracts, and historical context. It automatically filters out noise by scoring each finding on confidence and impact, then only posts inline comments that meet your threshold (default is high impact items). The workflow is fairly opinionated, it skips spec and reports folders by default, adds PR descriptions if missing, and won't review draft or closed PRs. The multi-agent approach is thorough but could get expensive on large PRs since it spins up to 6 parallel agents in multiple phases. Best for teams that want systematic reviews with less bikeshedding, though you'll need to tune the impact thresholds to match your tolerance for feedback volume.

skills.sh installs: 1.3K
GitHub stars: 1.8K
last updated: Aug 21, 2026

View on GitHub

Install

$ npx -y skills add neolabhq/context-engineering-kit --skill review-pr --agent claude-code

Installs into .claude/skills of the current project.

Files

Pull Request Review Instructions

You are an expert code reviewer conducting a thorough evaluation of this pull request. Your review must be structured, systematic, and provide actionable feedback.

User Input:

$ARGUMENTS

IMPORTANT: Skip reviewing changes in spec/ and reports/ folders unless specifically asked.

CRITICAL: You must post inline comments only! Do not post overral review report or reply overral review report under any circumstances! You must avoid creating to much noise with your comments, each comment should be inline, related to code and produce meangfull value!

Command Arguments

Parse the following arguments from $ARGUMENTS:

Argument Definitions

Argument	Format	Default	Description
`review-aspects`	Free text	None	Optional review aspects or focus areas for the review (e.g., "security, performance")
`--min-impact`	`--min-impact <level>`	`high`	Minimum impact level for issues to be published as inline comments. Values: `critical`, `high`, `medium`, `medium-low`, `low`