review-pr

Review changed paths in a pull request and post a confidence-gated routing comment.

3|Updated Jan 25, 2026
One-click install
npx skills add https://github.com/kapilvirenahuja/garura --skill review-pr-kapilvirenahuja
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: review-pr
Source: https://github.com/kapilvirenahuja/garura/tree/main/core/components/plays/review-pr
Command: npx skills add https://github.com/kapilvirenahuja/garura --skill review-pr-kapilvirenahuja

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

review-pr removes uncertainty in pull request reviews by performing a deterministic, diff-scoped quality assessment and applying a confidence-gated routing decision that can halt the ship pipeline on critical issues.

Core Features & Use Cases

  • Diff-scoped quality review: Evaluates only the changed paths and never runs a full-repo quality check.
  • Deterministic evidence-backed findings: Classifies findings using a PR severity taxonomy mapped from existing KB standard IDs.
  • Confidence-gated routing & reviewer selection: Computes a weighted confidence score, selects top reviewers from recent committers on changed paths (excluding the PR author), and posts/updates a structured routing comment.

Use case: When a team wants enterprise-grade PR quality gates, review-pr can automatically request changes (for P1) and escalate when confidence is below a configured threshold, while leaving a consistent, structured audit trail.

Quick Start

Run the review-pr skill for PR #123 and let it post a structured routing comment with confidence-gated outcomes.

Frequently Asked Questions about review-pr

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

FAQPage Schema
How do I automate a diff-scoped pull request quality review on GitHub?

A diff-scoped pull request quality review evaluates only the changed paths in a GitHub PR, classifies findings using a severity taxonomy, and posts a structured routing comment with a confidence score to gate the ship pipeline.

How does confidence scoring work for PR quality gates?

Confidence scoring for PR quality gates computes a weighted score from deterministic findings mapped to a PR severity taxonomy, halting the ship pipeline when confidence falls below a configured threshold and routing is set to block or escalate.

How do I automatically select code reviewers from recent GitHub committers?

Reviewer selection from recent committers analyzes activity on changed paths within a configured activity window, identifies top contributors excluding the PR author, and posts their names in a structured routing comment on the pull request.

Does this code review approach prevent full-repo quality checks from running?

Yes, this code review approach enforces constraints that prohibit reading outside the diff scope or invoking full-repo quality checks, ensuring the PR quality gate remains focused exclusively on changed paths.

When should I use a deterministic PR review instead of manual code review?

Use a deterministic PR review when you need enterprise-grade quality gates that automatically request changes for P1 findings, escalate low-confidence issues, and leave a consistent audit trail without subjective human judgment.

What configuration is needed to set up confidence-gated PR routing?

Confidence-gated PR routing requires configuration-driven standards paths, confidence thresholds, reviewer counts, activity windows for committer selection, and a bypass flag to control ship-pipeline halting behavior.