Evaluate code changes for quality, correctness, and adherence to project standards. Use when reviewing pull requests, verifying implementations against acceptance criteria, or auditing code for best practices.
Scanned 9/6/2026
Install to Claude Code
npx -y skills add aiFabricoCom/fabrico-collections --skill examples --agent claude-codeInstalls into .claude/skills of the current project.
Are you the author of Examples?
Add the live security badge to your README — it updates automatically with every re-scan.
[](https://www.skillsdirectory.com/skills/aifabricocom-examples)More formats (shields.io, HTML) on the badges page.
---
name: reviewing-code
description: "Evaluate code changes for quality, correctness, and adherence to project standards. Use when reviewing pull requests, verifying implementations against acceptance criteria, or auditing code for best practices."
---
# Reviewing Code
Evaluates code changes against project conventions, correctness, and maintainability. Focuses on actionable findings rather than stylistic nitpicks.
<principles>
<constructive-feedback>
Every finding must include a concrete suggestion. Identifying a problem without proposing a fix adds noise, not value.
</constructive-feedback>
<severity-over-volume>
Prioritize correctness and security issues over style preferences. A single missed null check matters more than ten formatting inconsistencies.
</severity-over-volume>
</principles>
## Review Process
Use the checklist below and track progress:
```
Progress:
- [ ] Step 1: Gather context
- [ ] Step 2: Analyze code changes
- [ ] Step 3: Produce review report
- [ ] Step 4: Validate findings
```
**Step 1: Gather context**
- Read the task description, acceptance criteria, and implementation plan if available.
- Identify the scope of changes — which files were added, modified, or deleted.
- Check project conventions by reading linter configs, existing patterns, and style guides.
**Step 2: Analyze code changes**
- Verify each change implements the stated requirements without unintended side effects.
- Check error handling, edge cases, and boundary conditions.
- Assess naming clarity, function length, and separation of concerns.
- Flag security risks: unvalidated input, exposed secrets, injection vectors.
**Step 3: Produce review report**
- Group findings by severity: critical, warning, suggestion.
- Each finding must reference the specific file and line range.
- Include a recommended fix or alternative approach for every issue raised.
**Step 4: Validate findings**
- Re-read each finding to confirm it is accurate and reproducible.
- Remove false positives and duplicate observations.
- Verify that critical findings are not masked by lower-severity noise.
## Connected Skills
- `codebase-analysing` - to understand project structure and patterns before reviewing
- `architecture-designing` - to evaluate architectural decisions in the reviewed code
</content>
Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.
No comments yet. Be the first to comment!