review-agent

Review code for bugs, vulnerabilities, and maintainability issues with file-level feedback.

Updated Dec 3, 2025
One-click install
npx skills add https://github.com/Unicorn/Radium --skill review-agent-unicorn
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: review-agent
Source: https://github.com/Unicorn/Radium/tree/main/skills/core/review-agent
Command: npx skills add https://github.com/Unicorn/Radium --skill review-agent-unicorn

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

The review-agent addresses the challenge of delivering fast, consistent, and actionable code reviews by identifying bugs, security vulnerabilities, style regressions, and testing gaps so teams can ship safer, higher-quality software.

Core Features & Use Cases

  • Bug & Vulnerability Detection: Pinpoint logic errors, insecure patterns, and potential exploit vectors in source code and pull requests.
  • Quality & Maintainability Guidance: Recommend refactors, naming improvements, and abstraction adjustments to reduce technical debt.
  • Testing & Coverage Advice: Assess unit and integration test coverage and suggest targeted tests for edge cases.
  • Performance and Best Practices: Highlight inefficient algorithms, N+1 query patterns, and resource management issues.
  • Use Case: A reviewer applies the agent to a pull request to get a structured review with critical security findings, major correctness issues, and a prioritized list of remediation steps.

Quick Start

Use the review-agent to analyze the repository path src/ and produce a structured code review report including critical issues, recommendations, and suggested tests.

Frequently Asked Questions about review-agent

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

FAQPage Schema
How do I automate code review for bugs and security vulnerabilities in a pull request?

Automating code review for bugs and security vulnerabilities in pull requests requires analyzing source code to pinpoint logic errors and insecure patterns. The review-agent applies this process to produce structured, actionable feedback with file-level locations and severity classifications.

What is the best way to identify N+1 query patterns and performance issues in source code?

Identifying N+1 query patterns and performance issues in source code requires highlighting inefficient algorithms and resource management problems. The review-agent applies these performance and best practice checks to generate prioritized remediation recommendations.

How do I assess test coverage and suggest targeted tests for edge cases?

Assessing test coverage and suggesting targeted tests for edge cases requires evaluating unit and integration test suites to find gaps. The review-agent provides testing and coverage advice to help developers build targeted tests for missed edge cases.

Can I use an automated review tool to reduce technical debt and improve code maintainability?

You can use an automated review tool to reduce technical debt and improve code maintainability by recommending refactors, naming improvements, and abstraction adjustments. The review-agent delivers quality and maintainability guidance to address these specific code issues.

Does the code review process work on individual code snippets or only full repositories?

The code review process works on both individual code snippets and full software repositories. The review-agent can analyze repository paths and apply its review logic to pull requests or specific code snippets to identify maintainability issues and bugs.