code-review

Review code changes against a plan and diff with six criteria.

Updated Mar 17, 2026
One-click install
npx skills add https://github.com/steven-3/supermind --skill code-review-steven-3
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: code-review
Source: https://github.com/steven-3/supermind/tree/main/skills/code-review
Command: npx skills add https://github.com/steven-3/supermind --skill code-review-steven-3

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

It ensures proposed changes are reviewed rigorously and consistently so defects, spec mismatches, missing tests, and security issues are caught before shipping.

Core Features & Use Cases

  • Structured Verify-Phase Review: Executes a step-by-step process that forces reviewers to read intent first, then read the full diff, then evaluate against all criteria.
  • Six-Criteria Issue Classification: Assesses Spec Compliance, Correctness, Test Coverage, Security, Maintainability, and Consistency with actionable output categories.
  • Decision-Grade Output: Produces a clear verdict (PASS | NEEDS FIXES | FAIL) with critical/important/suggestion sections and location-aware findings.
  • Implementer Feedback Protocol: Provides rules to prevent performative agreement and to evaluate findings thoughtfully when responding to review feedback.

Quick Start

Use the code-review skill during the Verify phase to review a feature implementation against its plan and diff, producing a PASS/NEEDS FIXES/FAIL verdict.

Frequently Asked Questions about code-review

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

FAQPage Schema
How do I catch security risks and spec mismatches in a pull request before merge?

Structured code review validates proposed changes against a plan and diff to catch security risks and spec mismatches before merge. It evaluates all changed files across six criteria to produce a release-gating verdict.

What criteria should I evaluate during a pre-merge code review?

Pre-merge code review should evaluate six criteria: Spec Compliance, Correctness, Test Coverage, Security, Maintainability, and Consistency. Each criterion generates actionable findings classified as Critical, Important, or Suggestion.

How do I enforce a structured verify-phase review for refactors and bug fixes?

Structured verify-phase review enforces reading intent first, then reading the full diff, then evaluating changes against all six criteria. It requires location-referenced findings to ensure traceability for refactors and bug fixes.

How do I classify code review findings to produce a release-gating verdict?

Classify code review findings as Critical, Important, or Suggestion with location-aware references to produce a release-gating verdict. The final output decision is categorized as PASS, NEEDS FIXES, or FAIL.

How should implementers respond to code review feedback without performative agreement?

Implementers should follow a feedback protocol that enforces thoughtful evaluation of review findings rather than performative agreement. This protocol prevents dismissive responses during the verify-phase gating process.

Can I use automated code review for security-sensitive code changes across typical PR workflows?

Automated code review applies to security-sensitive code changes across typical PR workflows by rigorously assessing diffs. It verifies test coverage and maintainability to output a clear PASS, NEEDS FIXES, or FAIL verdict.