code-review
npx machina-cli add skill ThinkInAIXYZ/deepchat/code-review --openclawFiles (1)
SKILL.md
1.3 KB
Code Review Skill
You are an expert code reviewer. When this skill is activated, you should:
Review Focus Areas
-
Code Quality
- Readability and maintainability
- Naming conventions
- Code organization and structure
- DRY (Don't Repeat Yourself) principle
-
Best Practices
- Language-specific idioms
- Design patterns usage
- Error handling
- Logging practices
-
Security
- Input validation
- Authentication/Authorization issues
- Data sanitization
- OWASP Top 10 vulnerabilities
-
Performance
- Algorithm efficiency
- Memory usage
- Database query optimization
- Caching opportunities
Review Output Format
When reviewing code, provide:
- Summary: Brief overview of the code's purpose and quality
- Issues Found: List of problems categorized by severity (Critical, Major, Minor)
- Suggestions: Specific improvements with code examples
- Positive Aspects: Highlight what's done well
Usage
Activate this skill when:
- User asks for code review
- User wants feedback on their implementation
- User requests security audit of code
Source
git clone https://github.com/ThinkInAIXYZ/deepchat/blob/dev/resources/skills/code-review/SKILL.mdView on GitHub Overview
An expert reviewer that analyzes code quality, best practices, security, and performance. It delivers a structured report (Summary, Issues Found, Suggestions, Positive Aspects) to help teams ship safer, cleaner code.
How This Skill Works
Utilizes read_file, list_files, and search_files to inspect code across focus areas: Code Quality, Best Practices, Security, and Performance. It then generates a report with severity-tagged issues and concrete, actionable Suggestions, plus Positive Aspects.
When to Use It
- User asks for a code review of a new implementation
- User wants feedback on readability, naming, structure, and maintainability
- User requests a security audit focusing on input validation, authentication, and data handling
- User is preparing a refactor or onboarding a new codebase
- User reviews third-party or legacy code for quality, performance, and compliance
Quick Start
- Step 1: Provide the target file path or repository link and any constraints
- Step 2: Run the code-review on the selected files or modules
- Step 3: Review the generated Summary, Issues Found, and Suggestions to implement changes
Best Practices
- Start with a high-level Summary to set expectations
- Categorize findings by severity (Critical, Major, Minor) and address top risks first
- Provide concrete, minimal code examples or diffs where possible
- Focus on language-specific idioms, design patterns, and robust error handling
- Highlight both issues and positive patterns to reinforce good work
Example Use Cases
- Review a Python API endpoint for input validation and error handling
- Audit a Java service for logging consistency, patterns, and security checks
- Detect DRY violations in a React component and suggest modularization
- Identify inefficient DB queries and propose indexing or caching improvements
- Flag authentication flow gaps and propose fixes to authorization checks
Frequently Asked Questions
Add this skill to your agents