Analyze code and return structured feedback with actionable items. Use when asked to review code, check quality, or identify issues in a file.
View on GitHubpotenlab/marketplace-potenlab
pl
skills/review/SKILL.md
January 25, 2026
Select agents to install to:
npx add-skill https://github.com/potenlab/marketplace-potenlab/blob/main/skills/review/SKILL.md -a claude-code --skill reviewInstallation paths:
.claude/skills/review/# Code Review Analyze the specified code and provide structured feedback. ## Target Review: $ARGUMENTS If no target specified, ask the user which file or directory to review. ## Review Checklist Evaluate the code against these criteria: ### Code Quality - [ ] Clear naming conventions - [ ] Appropriate function/method size - [ ] Single responsibility principle - [ ] No code duplication (DRY) ### Error Handling - [ ] Edge cases handled - [ ] Errors caught and handled appropriately - [ ] Input validation present ### Maintainability - [ ] Code is readable without excessive comments - [ ] Complex logic is documented - [ ] No magic numbers or strings ### Performance - [ ] No obvious performance issues - [ ] Appropriate data structures used - [ ] No unnecessary computations ## Output Format Provide your review in this exact structure: ### Summary [2-3 sentence overview of code quality and main findings] ### Issues Found | Severity | Location | Issue | Suggestion | |----------|----------|-------|------------| | [HIGH/MEDIUM/LOW] | [file:line] | [description] | [how to fix] | If no issues found, state: "No significant issues found." ### Strengths - [Positive observation about the code] - [Another positive observation] ### Actionable Items 1. [ ] [Specific action item with file:line reference] 2. [ ] [Another action item] If no actions needed, state: "No immediate actions required." ## Important - This is a READ-ONLY review. Do not modify any files. - Focus on actionable feedback, not stylistic preferences. - Severity levels: HIGH = bugs/security, MEDIUM = maintainability, LOW = suggestions.