code-review

Review pull requests and branch diffs with multi-agent analysis.

2|Updated Mar 16, 2026
One-click install
npx skills add https://github.com/hamzaPixl/pixl-ai --skill code-review-hamzapixl
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: code-review
Source: https://github.com/hamzaPixl/pixl-ai/tree/main/packages/crew/skills/code-review
Command: npx skills add https://github.com/hamzaPixl/pixl-ai --skill code-review-hamzapixl

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Multi-agent, structured PR review reduces missed defects, inconsistent feedback, and slow human review cycles by producing high-confidence, deduplicated findings and clear suggested actions before merge.

Core Features & Use Cases

  • Parallel specialist reviewers: Run 3-reviewer quick mode or 8-reviewer full mode across correctness, security, conventions, performance, API contracts, data migration, maintainability, and adversarial lenses.
  • Confidence scoring and classification: Deduplicate findings, boost consensus confidence, and label issues as AUTO-FIX for safe mechanical fixes or ASK for user decisions.
  • Optional auto-fix and PR posting: Apply low-risk fixes automatically and optionally post structured PR review comments via the GitHub CLI for seamless team workflows.
  • Use Case: Audit a branch or PR to identify security regressions, migration risks, and maintainability concerns before merge, or run automated periodic reviews on long-lived branches.

Quick Start

Review pull request 123 in full mode and post the findings as PR comments.

Frequently Asked Questions about code-review

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

FAQPage Schema
How do I automate pull request reviews to catch security and correctness issues before merge?

Automated pull request review uses a multi-agent system to analyze branch diffs, identifying security, correctness, and maintainability issues. It deduplicates findings with confidence scores and classifies them for auto-fixing or manual review before merge.

Can I automatically post code review comments to a GitHub pull request?

Yes, you can automatically post code review comments to a GitHub pull request. The review process produces structured findings and optionally posts them directly via the GitHub CLI for seamless team workflows.

What is the best way to review large branch diffs for data migration and API contract risks?

Reviewing large branch diffs for data migration and API contract risks is best handled by a full-mode multi-agent review. It deploys eight specialist reviewers to analyze adversarial issues, performance, and API contracts, producing high-confidence, deduplicated findings.

Does multi-agent code review support automatic fixing of pull request defects?

Yes, multi-agent code review supports automatic fixing of pull request defects. Findings are classified as AUTO-FIX for safe mechanical fixes or ASK for user decisions, allowing low-risk issues to be applied automatically.

When do I need a quick mode versus a full mode for branch diff analysis?

You need quick mode with three reviewers for routine branch diff analysis and full mode with eight specialist reviewers for large or critical changes. Both modes deduplicate findings and provide confidence scores for the detected issues.

How does multi-agent code review handle conflicting feedback from different reviewers?

Multi-agent code review handles conflicting feedback by deduplicating findings and boosting consensus confidence across specialist reviewers. This process reduces inconsistent feedback and produces clear suggested actions for each identified issue.