/review
Perform a thorough code review of the current uncommitted changes or a specified file/directory.
$ npx -y skills add rohitg00/awesome-claude-code-toolkit --agent claude-codeHow it fires
How this command gets triggered: by you, by Claude, or both.
- Fires itselfClaude auto-loads it when your prompt matches the work.
- You can call itInvoke it directly when you want it.
- Slash command
/review
Context preview
What this command does when you run it.
Perform a thorough code review of the current uncommitted changes or a specified file/directory.
Command definition
review.mdname: review
description: Perform a thorough code review of the current uncommitted changes or a specified file/directory.
/code-guardian:review
Perform a thorough code review of the current uncommitted changes or a specified file/directory.
Process
1. Gather the changes to review:
- Run `git diff` for unstaged changes and `git diff --cached` for staged changes
- If a specific file or directory is provided as an argument, focus the review there
- If no changes exist, review the most recent commit with `git diff HEAD~1`
2. Analyze each changed file across these dimensions:
Correctness
- Logic errors, off-by-one mistakes, incorrect boolean conditions
- Unhandled null/undefined values and missing error boundaries
- Race conditions in async code, missing awaits, unhandled promise rejections
- Incorrect type assertions or unsafe casts
Security
- User input flowing to SQL queries, shell commands, or file paths without sanitization
- Hardcoded credentials, API keys, tokens, or connection strings
- Missing authentication or authorization checks on new endpoints
- Unsafe deserialization of external data
Performance
- N+1 query patterns in database access
- Unnecessary re-renders in React components (missing memoization, unstable references)
- Unbounded data fetching without pagination or limits
- Synchronous blocking operations on hot paths
- Memory leaks from unclosed resources, event listeners, or subscriptions
Design
- Single Responsibility violations: functions or classes doing too many things
- Tight coupling between modules that should be independent
- Missing abstractions where patterns repeat three or more times
- Public API surface area: are new exports intentional and well-designed?
Readability
- Variable and function names that obscure intent
- Deeply nested conditionals that should be flattened or extracted
- Magic numbers or strings that should be named constants
- Missing or misleading type annotations
3. Classify each finding by severity:
- **CRITICAL** - Will cause bugs, data loss, or security vulnerabilities in production
- **WARNING** - Likely to cause issues under certain conditions or during maintenance
- **SUGGESTION** - Improvement opportunity that is not urgent
4. For each finding, provide:
- The file path and line range
- A clear description of the issue
- A concrete fix or code suggestion
Output Format
Present findings grouped by severity, then by file. Lead with critical issues. Skip categories that have no findings. End with a brief summary of overall quality.
Rules
- Be specific and actionable, not vague ("this could be improved")
- Do not flag style preferences unless they impact readability significantly
- Acknowledge well-written code briefly at the end
- Limit total findings to the 15 most impactful to avoid noise
- If the diff is very large, prioritize new code over modified code
Read more
name: review description: Perform a thorough code review of the current uncommitted changes or a specified file/directory.
/code-guardian:review
Perform a thorough code review of the current uncommitted changes or a specified file/directory.
Process
1. Gather the changes to review:
- Run `git diff` for unstaged changes and `git diff --cached` for staged changes
- If a specific file or directory is provided as an argument, focus the review there
- If no changes exist, review the most recent commit with `git diff HEAD~1`
2. Analyze each changed file across these dimensions:
Correctness
- Logic errors, off-by-one mistakes, incorrect boolean conditions
- Unhandled null/undefined values and missing error boundaries
- Race conditions in async code, missing awaits, unhandled promise rejections
- Incorrect type assertions or unsafe casts
Security
- User input flowing to SQL queries, shell commands, or file paths without sanitization
- Hardcoded credentials, API keys, tokens, or connection strings
- Missing authentication or authorization checks on new endpoints
- Unsafe deserialization of external data
Performance
- N+1 query patterns in database access
- Unnecessary re-renders in React components (missing memoization, unstable references)
- Unbounded data fetching without pagination or limits
- Synchronous blocking operations on hot paths
- Memory leaks from unclosed resources, event listeners, or subscriptions
Design
- Single Responsibility violations: functions or classes doing too many things
- Tight coupling between modules that should be independent
- Missing abstractions where patterns repeat three or more times
- Public API surface area: are new exports intentional and well-designed?
Readability
- Variable and function names that obscure intent
- Deeply nested conditionals that should be flattened or extracted
- Magic numbers or strings that should be named constants
- Missing or misleading type annotations
3. Classify each finding by severity:
- **CRITICAL** - Will cause bugs, data loss, or security vulnerabilities in production
- **WARNING** - Likely to cause issues under certain conditions or during maintenance
- **SUGGESTION** - Improvement opportunity that is not urgent
4. For each finding, provide:
- The file path and line range
- A clear description of the issue
- A concrete fix or code suggestion
Output Format
Present findings grouped by severity, then by file. Lead with critical issues. Skip categories that have no findings. End with a brief summary of overall quality.
Rules
- Be specific and actionable, not vague ("this could be improved")
- Do not flag style preferences unless they impact readability significantly
- Acknowledge well-written code briefly at the end
- Limit total findings to the 15 most impactful to avoid noise
- If the diff is very large, prioritize new code over modified code
The most comprehensive toolkit for Claude Code -- 135 agents, 35 curated skills (+400,000 via SkillKit), 42 commands, 176+ plugins, 20 hooks, 15 rules, 7 templates, 15 MCP configs, 26 companion apps, 53 ecosystem entries, and more.
Repo: rohitg00/awesome-claude-code-toolkit
Other commands on rohitg00-claude-code-toolkit.
- /adr
Write an Architecture Decision Record documenting a significant technical decision.
Open command - /design-review
Conduct a structured design review of a module, feature, or system component.
Open command - /diagram
Generate Mermaid diagrams from codebase analysis or description.
Open command - /migrate
Plan and execute a framework or library migration incrementally.
Open command - /plan
Create a structured implementation plan for the requested feature or change.
Open command - /refactor
Perform a systematic refactoring of the specified code area.
Open command

