- Home
- Skills
- 404kidwiz
- Claude Supercode Skills
- Code Reviewer Skill
code-reviewer-skill_skill
- Python
21
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 404kidwiz/claude-supercode-skills --skill code-reviewer-skill- SKILL.md3.1 KB
Overview
This skill is an expert code reviewer focused on correctness, security, performance, and maintainability. It inspects code changes and pull requests to identify bugs, vulnerabilities, and quality issues while suggesting clear, actionable fixes. Use it to gate merges, improve test coverage, and harden code before deployment.
How this skill works
I read the PR description and diffs, then prioritize findings by severity: security and correctness first, performance and reliability next, and style or documentation as low-priority suggestions. I surface concrete examples, minimal reproduction steps when relevant, and recommended code changes or test additions. Automated tool findings (linters, SAST) are integrated into the review when available.
When to use it
- Reviewing pull requests or proposed code changes before merge
- Performing security audits or threat checks on code
- Identifying potential bugs and logic errors early
- Assessing performance and resource usage implications
- Validating error handling, input validation, and edge cases
Best practices
- Focus reviews on code and behavior, not the author
- Prioritize feedback by impact: security and correctness first
- Give clear, actionable fixes and show the rationale
- Use automated linters/SAST before manual review to reduce noise
- Acknowledge good patterns and improvements, not just problems
Example use cases
- Blocking a merge because of an authentication bypass or SQL injection
- Pointing out an N+1 query in a recent change and suggesting a batched query or eager load
- Flagging hardcoded secrets or improper cryptographic usage and advising remediation
- Requesting additional unit or integration tests for uncovered edge cases
- Identifying missing input validation and providing example sanitization
FAQ
No. It recommends minimal, targeted changes and refactors only when essential; large refactors are suggested as future work.
How does it treat style issues?
Style issues are noted as non-blocking suggestions; prefer automated formatters to enforce style rules.