Pr Review Expert

PR Review Expert is an AI skill that provides comprehensive pull request review covering code quality, architecture alignment, testing adequacy, and deployment readiness. It covers diff analysis, change impact assessment, review comment generation, approval criteria, and review workflow optimization that produce thorough, constructive pull request reviews.

What Is This?

Overview

PR Review Expert delivers systematic pull request review processes that evaluate changes across multiple quality dimensions. It addresses diff analysis that understands the scope of changes, code quality assessment for readability and team standards adherence, architecture alignment checking, test coverage evaluation, deployment risk assessment, and structured feedback with severity levels.

Who Should Use This

This skill serves developers reviewing teammates' pull requests, team leads establishing review standards and checklists, senior engineers reviewing complex changes across multiple services, and teams implementing consistent review practices.

Why Use It?

Problems It Solves

Pull request reviews vary wildly in depth and quality depending on the reviewer, the time available, and the size of the change. Large PRs receive superficial reviews because thorough review is exhausting. Reviewers focus on style issues while missing logic bugs or security vulnerabilities. Without structured criteria, review quality is unpredictable.

Core Highlights

The skill evaluates PRs across defined quality dimensions so nothing important is missed. Severity classification helps authors prioritize feedback. Constructive feedback includes explanations and alternative approaches rather than just identifying problems. Review summaries provide clear approve, request changes, or comment recommendations.

How to Use It?

Basic Usage

class PRReviewChecklist:
    dimensions = [
        {"name": "correctness", "checks": [
            "Logic handles all expected input cases",
            "Edge cases are addressed",
            "Error handling covers failure modes"
        ]},
        {"name": "security", "checks": [
            "Input validation present",
            "No sensitive data in logs",
            "Authentication checks where required"
        ]},
        {"name": "testing", "checks": [
            "New code has corresponding tests",
            "Tests cover both success and failure paths",
            "No test logic depends on external state"
        ]},
        {"name": "architecture", "checks": [
            "Changes follow existing patterns",
            "No unnecessary coupling introduced",
            "Public API changes are backward compatible"
        ]}
    ]

    def review(self, diff):
        findings = []
        for dim in self.dimensions:
            for check in dim["checks"]:
                result = self.evaluate(diff, check)
                if not result["passes"]:
                    findings.append({"dimension": dim["name"],
                        "check": check, "severity": result["severity"]})
        return findings

Real-World Examples

PR Review Summary
PR #312: Implement order cancellation API
Author: @developer | Files: 8 | Lines: +247 / -12

Overall: Request Changes (1 blocking issue)

[CRITICAL] src/api/orders.py:89
Missing authorization check. Any authenticated user can cancel
any order. Should verify request.user.id matches order.customer_id.
Suggested fix:
  if order.customer_id != request.user.id:
      raise ForbiddenError("Not authorized to cancel this order")

[WARNING] src/services/order_service.py:145
Cancellation does not check if order is already shipped.
Orders in "shipped" status should not be cancellable.

[SUGGESTION] src/api/orders.py:95
Consider returning the updated order object in the response
so the client does not need a separate GET request.

[GOOD] tests/test_order_cancellation.py
Test coverage is comprehensive with 8 test cases covering
happy path, authorization, and edge cases.

Dimension Scores:
  Correctness: Needs work (missing status check)
  Security: Blocking (missing authorization)
  Testing: Good
  Architecture: Good (follows existing patterns)

Advanced Tips

Review the test changes first to understand the intended behavior before reading the implementation. For large PRs, start with the API or interface changes to understand the contract, then review the implementation. Keep a personal checklist of common issues you find to build pattern recognition over time.

When to Use It?

Use Cases

Use PR Review Expert when reviewing pull requests and wanting a structured approach, when establishing review standards for a development team, when training junior developers on effective code review practices, or when reviewing large or complex PRs that need thorough analysis.

Related Topics

Code review best practices, GitHub and GitLab review workflows, static analysis tools, automated review bots, and pair programming as a review alternative all complement PR review processes.

Important Notes

Requirements

Access to the pull request diff and related context like linked issues. Understanding of the project's coding standards and architecture patterns. Time allocated for thorough review proportional to the size of the change.

Usage Recommendations

Do: provide specific, actionable feedback with line references and suggested fixes. Acknowledge good code and patterns alongside issues to maintain positive team dynamics. Review both the code and the tests to ensure tests actually validate the intended behavior.

Don't: block PRs on style preferences not codified in team standards. Leave review comments without explaining why the change matters. Rush reviews to unblock merges, as superficial reviews provide false confidence in code quality.

Limitations

Structured review catches pattern-based issues but may miss domain-specific logic errors that require deep business knowledge. Large PRs reduce review effectiveness regardless of the review process used. Automated review tools complement but cannot fully replace human judgment on design decisions and code clarity.