review-code-engineer

Automate repository-aware code reviews and output actionable findings with file paths and risk levels.

2|1|Updated Sep 11, 2025
One-click install
npx skills add https://github.com/ThiagoGuislotti/copilot-instructions --skill review-code-engineer
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: review-code-engineer
Source: https://github.com/ThiagoGuislotti/copilot-instructions/tree/main/.codex/skills/review-code-engineer
Command: npx skills add https://github.com/ThiagoGuislotti/copilot-instructions --skill review-code-engineer

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Perform repository-aware code reviews focused on bugs, regressions, architectural violations, missing tests, and operational risks. Use when the user asks for review of diffs, pull requests, patches, or implementation quality.

Core Features & Use Cases

  • Load minimal context first: Load .github/AGENTS.md and .github/copilot-instructions.md, then .github/instructions/pr.instructions.md, and route additional domain instructions via .github/instruction-routing.catalog.yml based on changed files.
  • Review order: Evaluate correctness, architecture boundary violations, missing/weak tests, security, resilience, and performance risks.
  • Output-driven findings: present findings with file path, risk, evidence, and precise fix recommendations; provide explicit residual risks if no findings.
  • Useful references: Refer to guided documents such as the provided .github instruction files and chat modes.

Quick Start

Provide the diff or PR you want reviewed, and I will assess correctness, architecture, tests, and operational risks.

Frequently Asked Questions about review-code-engineer

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

FAQPage Schema
How do I automate code review for a pull request to catch bugs and regressions?

Repository-aware code review automates pull request analysis to catch bugs, regressions, and architectural violations. It evaluates diffs against repository context, outputting findings with file paths, risk levels, evidence, and precise fix recommendations for targeted remediation.

What is repository-aware diff review and how does it work?

Repository-aware diff review loads minimal context like agent instruction files and domain routing catalogs before evaluating changed files. It assesses correctness, architecture boundaries, missing tests, security, resilience, and performance risks to surface actionable findings across the codebase.

Can I use automated code review to check for architectural violations and missing tests?

Yes, automated code review specifically evaluates architectural boundary violations and missing or weak tests alongside correctness and security risks. It routes domain instructions based on changed files and provides actionable findings with evidence and precise fix recommendations.

How do I assess operational and security risks in a code patch?

Provide the diff or pull request for review to assess operational and security risks in a code patch. The evaluation checks resilience and security vulnerabilities, outputting findings with risk levels, file paths, evidence, and explicit residual risks if no issues are found.

Does automated code review work with custom repository instructions and routing catalogs?

Yes, automated code review loads `.github/AGENTS.md`, copilot instructions, and PR instructions first, then routes additional domain instructions via a catalog YAML file based on the specific changed files in the diff.

What is the best way to get actionable fix recommendations from a code review?

Provide a diff or pull request for repository-aware review to get actionable fix recommendations. It outputs findings with precise fix recommendations, file paths, risk levels, and evidence, enabling targeted remediation of bugs and architectural violations.