skills.rest
usmangurowa
usmangurowaCOMMUNITY
Explore all skills by this author →
Website𝕏@UsmanGurowaGitHub

review-pr

Reviews pull requests using specialized reviewer personas with cross-review and debate.

Updated Jul 17, 2026
Search Tags:#code review#pull request#claude code#codex#multi-agent#debate#reviewer personas
One-click install
npx skills add https://github.com/usmangurowa/socialmocks --skill review-pr-usmangurowa
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: review-pr
Source: https://github.com/usmangurowa/socialmocks/tree/main/.agents/skills/review-pr
Command: npx skills add https://github.com/usmangurowa/socialmocks --skill review-pr-usmangurowa

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Pull request reviews are often shallow, inconsistent, or bottlenecked on a single reviewer's perspective. This Skill orchestrates a team of specialized reviewer personas that examine a PR from architecture, code quality, testing, security, and product angles, then debate findings before delivering a unified review. ## Core Features & Use Cases - Multi-Persona Review: Specialized reviewers (Architecture Skeptic, Code Quality, Test, Security & Reliability, Product) each evaluate the PR through their own lens. - Portable Orchestration: Runs in parallel when the runtime supports multi-agent orchestration (Claude Code teams, Codex sub-agents, opencode subagents), or falls back to sequential lead-mediated passes. - Cross-Review and Debate: Findings are challenged and validated across reviewers before being folded into a single human-readable summary. - Use Case: Before merging a feature branch, invoke this Skill to get an end-to-end review covering correctness, edge cases, test coverage, UX impact, and whether the PR solves the right problem. ## Quick Start Ask the AI to review the current pull request using the review-pr skill and produce a unified review summary.

Frequently Asked Questions about review-pr

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

FAQPage Schema
How do I run a multi-agent code review on a pull request?▼

Invoke the review-pr skill on your pull request and it dispatches specialized reviewer personas in parallel when the runtime supports multi-agent orchestration. Otherwise the lead reviewer runs the same roles sequentially and merges the findings.

What reviewer perspectives does an AI PR review cover?▼

The review covers architecture soundness, code quality, test coverage of critical logic, security and reliability concerns, and product/UX impact. Each persona applies its own lens and findings are debated before the final summary.

Does this PR review work without multi-agent support?▼

Yes. When the runtime lacks multi-agent orchestration, the lead runs each reviewer role as lead-mediated subagent jobs or sequential passes, producing the same cross-reviewed output.

Which AI coding tools can orchestrate this PR review?▼

The skill supports Claude Code teams, Codex sub-agents, and opencode subagents for parallel orchestration, with a sequential fallback for other runtimes.

When should I not use a team-based PR review?▼

Avoid it for trivial changes like typo fixes where a full multi-persona review adds overhead without value. It is designed for PRs where correctness, edge cases, and design decisions genuinely matter.

Related Skills

Explore More Software Engineering Skills→
n8n-ion8n-io
199.9k

create-pr

Create GitHub pull requests with compliant type(scope): summary titles via CLI or API.

Official
Advanced
badbsallyybadbsallyy

prepare-pr

Rebase a GitHub pull request onto main and run automated checks.

Community
Advanced
Open-PawsOpen-Paws
3

code-review

Automate layered code review for pull requests in CI pipelines.

Official
Advanced
bordenetbordenet
8

micro-harsh-review

Scores code changes with three-critic review and flags anti-patterns before commit.

Community
Intermediate