pr-review

Aggregate Copilot, SonarQube, CI, and agent findings into a tiered GitHub PR report.

Updated Nov 24, 2025
One-click install
npx skills add https://github.com/ByronWilliamsCPA/.claude --skill pr-review-byronwilliamscpa
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: pr-review
Source: https://github.com/ByronWilliamsCPA/.claude/tree/main/.claude/skills/pr-review
Command: npx skills add https://github.com/ByronWilliamsCPA/.claude --skill pr-review-byronwilliamscpa

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Pull request review and remediation is often fragmented across multiple tools, requiring manual coordination between Copilot, SonarQube, CI checks, and human reviewers. This Skill unifies the entire process into two slash commands that automatically gather findings from every source and resolve them in an isolated worktree.

Core Features & Use Cases

  • Tiered Review: The pr-review command triggers Copilot, SonarQube, and 8 parallel agents to produce a prioritized findings report with Critical, Important, Suggested, and Informational tiers.
  • Automated Remediation: The pr-fix command gathers all open PR issues including CI failures, review comments, SonarQube findings, and agent findings, then fixes them in an isolated worktree without contaminating the main working tree.
  • Use Case: A team member opens a PR with failing CI checks, Copilot suggestions, and SonarQube security hotspots. Running pr-review produces a unified report, and pr-fix automatically resolves all fixable issues while flagging design debates for human input.

Quick Start

Use the pr-review skill to review and fix issues on a GitHub pull request by providing its URL.

Frequently Asked Questions about pr-review

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

FAQPage Schema
How do I automate GitHub pull request reviews with Copilot and SonarQube findings?

Automated pull request review aggregates Copilot, SonarQube, and CI check findings into a unified tiered report. It uses parallel AI agents to categorize issues by severity and flags design debates for human input before merge.

Can I automatically fix failing CI checks and code review comments in a GitHub PR?

Automated PR remediation gathers all open issues including CI failures and review comments, then fixes them automatically. It applies fixes in an isolated git worktree to prevent contaminating the main working tree during the process.

Do I need GitHub CLI authentication and SonarQube configuration to run automated PR reviews?

Yes, GitHub CLI authentication and SonarQube project configuration are required. The automated review process depends on these tools to aggregate findings, and an isolated git worktree is needed to apply fixes safely.

What is the best way to consolidate security hotspots and code quality issues before merging a pull request?

Consolidating security hotspots and code quality issues requires a tiered review approach. Running a unified review command triggers multiple parallel agents to produce a prioritized report with Critical, Important, Suggested, and Informational tiers.

How does an isolated git worktree prevent contamination when applying automated PR fixes?

An isolated git worktree separates automated fixes from the main working tree. Remediation commands apply changes within this isolated environment, preventing contamination and ensuring the primary repository state remains stable.

Why does automated PR remediation flag certain SonarQube findings for human input instead of fixing them?

Automated remediation flags design debates and complex findings for human input because they require contextual decision-making. It only applies fixes to resolvable issues like CI failures and standard code quality suggestions automatically.