code-review

Automate structured code reviews with four-layer analysis and file:line references.

Updated Mar 5, 2026
One-click install
npx skills add https://github.com/Xerrion/opencode --skill code-review-xerrion
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: code-review
Source: https://github.com/Xerrion/opencode/tree/main/skills/code-review
Command: npx skills add https://github.com/Xerrion/opencode --skill code-review-xerrion

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Comprehensive, repeatable code reviews that systematically uncover defects, security vulnerabilities, performance bottlenecks, and maintainability issues before changes are merged.

Core Features & Use Cases

  • Four-layer review framework (Correctness, Security, Performance, Style & Maintainability) to guide thorough analysis.
  • Severity classification (Critical, Major, Minor, Nitpick) with actionable remediation guidance.
  • Confidence threshold requiring findings to have at least 80% certainty before reporting.
  • Structured review workflow including initial scan, deep analysis, context evaluation, philosophy compliance, synthesis, and clear file:line references.
  • Use Cases: before reporting implementation completion; during explicit review requests; as an independent audit after code changes; with review commands and pipelines.

Quick Start

Initiate an end-to-end code-review pass on the latest changes using the four layers and report findings with file:line references.

Frequently Asked Questions about code-review

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

FAQPage Schema
How do I automate a structured pull request review process?

Automate a structured pull request review by applying a four-layer model—Correctness, Security, Performance, and Style—to systematically uncover defects and vulnerabilities before merging code changes.

What is the best way to perform a security and compliance audit after code changes?

Conduct a post-commit compliance audit by running an independent review pass that scans for security vulnerabilities and maintainability issues, classifying severity and providing actionable remediation guidance.

How does confidence threshold work in static analysis code reviews?

Confidence thresholds in static analysis code reviews filter out false positives by requiring findings to reach at least 80% certainty before they are reported, ensuring high-quality remediation guidance.

Can I use this code review process across multiple languages and frameworks?

Yes, the code review process applies across multiple languages and frameworks during pre-merge reviews and post-commit audits, requiring no specific dependencies to analyze varied software projects.

What is the four-layer code review framework?

The four-layer code review framework evaluates software changes across Correctness, Security, Performance, and Style & Maintainability to guide thorough analysis and identify bottlenecks.

How do I ensure code review findings are actionable?

Ensure actionable code review findings by enforcing mandatory file:line references and classifying severity levels—Critical, Major, Minor, and Nitpick—within a structured review workflow synthesis.