skill-code-review

Synthesize multi-LLM code review findings into inline PR comments.

1|Updated Jun 12, 2026
One-click install
npx skills add https://github.com/mhdxbilal/Ai --skill skill-code-review-mhdxbilal
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: skill-code-review
Source: https://github.com/mhdxbilal/Ai/tree/main/.claude/skills/skill-code-review
Command: npx skills add https://github.com/mhdxbilal/Ai --skill skill-code-review-mhdxbilal

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Coordinating high-velocity, multi-provider code reviews is complex and time-consuming; this skill orchestrates multiple AI reviewers to produce a unified, actionable assessment with inline PR feedback.

Core Features & Use Cases

  • Orchestrated Multi-LLM Review: coordinates several AI reviewers to surface diverse perspectives on code quality, security, and architecture.
  • Inline PR Comments & Synthesis: generates concrete inline comments and a consolidated synthesis suitable for PR discussions.
  • Compliance Gates & Risk Signals: enforces defined validation gates, highlights critical risks, and suggests remediation steps.
  • Use Case: when reviewing a pull request with security and quality implications, run the skill to obtain a comprehensive, multi-model analysis.

Quick Start

Provide a pull request or code diff to start the full multi-LLM code-review pipeline and receive a structured synthesis.

Frequently Asked Questions about skill-code-review

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

FAQPage Schema
How do I automate multi-LLM code reviews on a pull request?

To automate multi-LLM code reviews, provide a pull request or code diff to orchestrate several AI reviewers, synthesizing their findings into inline PR comments and a consolidated risk assessment.

Can I use multiple AI models to find security and quality issues in a code diff?

Yes, coordinating multiple AI models on a code diff surfaces diverse perspectives on security and quality issues, synthesizing the findings into a unified assessment with actionable inline PR feedback.

What is the best way to synthesize AI code review feedback from different providers?

The best way to synthesize AI code review feedback is using a structured pipeline that aggregates findings from multiple providers, enforcing compliance gates to produce verifiable, consolidated outputs.

How do I enforce validation gates and compliance checks during automated code review?

Automated code review enforces validation gates and compliance checks by applying structured review phases to pull requests, highlighting critical risks and suggesting remediation steps before outputs are synthesized.

Does multi-LLM code review work on local code directories or only pull request diffs?

Multi-LLM code review works on both local code directories and pull request diffs, analyzing the provided source to surface security, quality, and architectural issues across the codebase.

When should I avoid using a multi-provider pipeline for code review?

Avoid using a multi-provider pipeline for code review when you lack explicit gating mechanisms to ensure verifiable outputs, as synthesizing raw findings without structured phases may produce unactionable feedback.