# Review Codes

> Review code changes for quality, design, and maintainability. Use when reviewing pull requests, code changes, or performing code reviews. Applies general engineering code review standards. Use when this capability is needed.

- Skill: `tomevault-io/review-codes` (Agent Skill, multi-file: 2 files)
- Install (CLI): `npx skillmds@latest add tomevault-io/review-codes`
- Raw SKILL.md: https://api.skillmd.com/api/skills/tomevault-io/review-codes/raw
- Safety review: pending (external: skill-scanner PASS, skillspector PASS)
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Coding & Dev Tools
- Author: tomevault-io (https://skillmd.com/u/tomevault-io)
- Updated: 2026-09-17
- Page: https://skillmd.com/skills/tomevault-io/review-codes

---


# Code Review Skill

Review code changes following the project's Code Review Process Guidelines.

## When to Use

Use this skill when:
- Reviewing pull requests or merge requests
- Performing code reviews on any codebase
- Evaluating code quality and design decisions
- Providing constructive feedback on code changes

## Instructions

### Step 1: Load the Guidelines

Read the code review process guidelines:
- **Primary**: `.aiassisted/guidelines/engineering/code-review-process.md`

### Step 2: Understand the Change

1. Read the change description (PR title, commit messages)
2. Ask: Does this change make sense? Is it solving the right problem?
3. If the change should not exist, explain why and suggest alternatives

### Step 3: Identify Main Components

Find the primary files with significant logical changes. Review these first to understand context.

**Communicate major design issues immediately** - don't wait until reviewing all files.

### Step 4: Review Systematically

Check each area from the guidelines:

| Area | What to Check |
|------|---------------|
| **Design** | Do interactions make sense? Right place in codebase? |
| **Functionality** | Does it work as intended? Edge cases? Concurrency issues? |
| **Complexity** | Is it more complex than necessary? Over-engineered? |
| **Testing** | Appropriate tests? Will they catch regressions? |
| **Naming** | Descriptive? Follows conventions? |
| **Comments** | Explain "why" not "what"? Accurate? |
| **Style** | Follows style guide? Consistent? |
| **Documentation** | README/API docs updated if needed? |

### Step 5: Provide Feedback

Format your review using severity labels:

```
**Nit:** Consider renaming `x` to `userCount` for clarity.

**Optional:** This could use early return to reduce nesting.

**FYI:** The new `async/await` syntax simplifies this pattern.

This function has a potential null pointer issue on line 42.

**Blocking:** This introduces a SQL injection vulnerability.
```

#### Severity Labels

| Label | Meaning |
|-------|---------|
| **Nit:** | Minor suggestion, optional |
| **Optional:** | Consider this, not required |
| **FYI:** | Educational, no action needed |
| *(no label)* | Should be addressed before approval |
| **Blocking:** | Must be fixed before merge |

### Step 6: Make a Decision

**Approve** when:
- Change improves overall code health
- Functions correctly and is well-tested
- Follows project standards
- No security vulnerabilities

**Request Changes** when:
- Change worsens code health
- Contains bugs or logic errors
- Missing tests or security issues
- Approach is fundamentally flawed

## Review Comment Guidelines

### Good Comments

- Focus on the code, not the developer
- Explain the reasoning and impact
- Reference best practices or documentation
- Point out problems; let developers find solutions

### Avoid

- "Why did you do it this way?" (judgmental)
- "This doesn't look right." (vague)
- "Use X instead." (prescriptive without explanation)

## Core Principle

> **Approve a change once it definitely improves the overall code health of the system, even if it is not perfect.**

Balance developer velocity with code quality. Reviews should be collaborative, not adversarial.

## Reference

The complete code review process is documented in:
`.aiassisted/guidelines/engineering/code-review-process.md`

Always load and follow the guidelines as the authoritative source.

---
> Converted and distributed by [TomeVault](https://tomevault.io/claim/rstlix0x0) — claim your Tome and manage your conversions.
<!-- tomevault:4.0:skill_md:2026-04-13 -->