- Home
- Skills
- Jeffallan
- Claude Skills
- Code Reviewer
code-reviewer_skill
- HTML
110
GitHub Stars
1
Bundled Files
2 months ago
Catalog Refreshed
4 months ago
First Indexed
Readme & install
Copy the install command, review bundled files from the catalogue, and read any extended description pulled from the listing source.
Installation
Preview and clipboard use veilstrat where the catalogue uses aiagentskills.
npx veilstrat add skill jeffallan/claude-skills --skill code-reviewer- SKILL.md3.0 KB
Overview
This skill acts as a senior principal engineer for code reviews, delivering focused, actionable feedback that improves correctness, security, performance, and maintainability. It combines a checklist-driven workflow with prioritized recommendations and concrete code examples to help teams ship safer, cleaner code. Use it to turn pull requests and audits into learning moments and reliable releases.
How this skill works
I first read the PR description and related context to understand the intent and scope. I inspect architecture, design choices, implementation details, tests, and security posture, then produce a structured review: summary, prioritized issues, positive notes, questions, and a clear verdict. Every recommendation includes examples or diffs where useful, and issues are categorized (critical, major, minor) to guide remediation. Tests and spec compliance are validated alongside code quality and common anti-patterns.
When to use it
- Reviewing pull requests before merge
- Conducting code quality or design audits
- Identifying security vulnerabilities and OWASP issues
- Spotting refactoring opportunities and technical debt
- Validating test coverage and behavior against specs
Best practices
- Always provide the problem context and reproduction steps when possible
- Prioritize feedback: block on security and correctness first, style last
- Include minimal, runnable code examples or diffs for suggested fixes
- Praise strong patterns and tested code to reinforce good practices
- Verify tests and edge cases rather than assuming coverage is sufficient
Example use cases
- PR review for a backend API change, checking data validation, auth, and query performance
- Audit a feature branch to highlight N+1 queries, magic numbers, or inefficient loops
- Security-focused review to identify injections, unsafe deserialization, or misconfigured auth checks
- Refactoring suggestions that decrease complexity while preserving behavior and improving testability
- Final report for a release candidate summarizing critical blockers and test gaps
FAQ
A concise summary, prioritized issues (critical→minor), positive feedback, questions for the author, and a verdict (approve/request changes/comment).
Will you rewrite my code for me?
I provide precise, minimal code examples and diffs to demonstrate fixes, but I do not replace the author—recommendations are designed for quick application by the developer.