Code Review

Code Review Skill

You are a code review assistant. When reviewing code, follow these steps:

Review Process

1. Check Style: Reference the style guide using get_skill_reference("code-review", "style-guide.md") 2. Run Style Check: Use get_skill_script("code-review", "check_style.py") for automated style checking 3. Look for Issues: Identify potential bugs, security issues, and performance problems 4. Provide Feedback: Give structured feedback with severity levels

Feedback Format

• Critical: Must fix before merge (security vulnerabilities, bugs that cause crashes)
• Important: Should fix, but not blocking (performance issues, code smells)
• Suggestion: Nice to have improvements (naming, documentation, minor refactoring)

Review Checklist

• [ ] Code follows naming conventions
• [ ] No hardcoded secrets or credentials
• [ ] Error handling is appropriate
• [ ] Functions are not too long (< 50 lines)
• [ ] No obvious security vulnerabilities
• [ ] Tests are included for new functionality

Python Style Guide

Naming Conventions

Variables and Functions

• Use snake_case for variables and functions
• Use descriptive names that explain the purpose
• Avoid single-letter names except for loop counters

# Good
user_count = 10
def calculate_total_price(items):
    pass

# Bad
uc = 10
def calc(i):
    pass

Classes

• Use PascalCase for class names
• Use nouns that describe what the class represents

# Good
class UserAccount:
    pass

class OrderProcessor:
    pass

# Bad
class user_account:
    pass

class Process:
    pass

Constants

• Use UPPER_SNAKE_CASE for constants
• Define at module level

# Good
MAX_RETRY_COUNT = 3
DEFAULT_TIMEOUT = 30

# Bad
maxRetryCount = 3
default_timeout = 30

Code Organization

Imports

• Group imports in this order: standard library, third-party, local
• Sort alphabetically within each group
• Use absolute imports

# Standard library
import os
import sys
from typing import List, Optional

# Third-party
import requests
from pydantic import BaseModel

# Local
from myapp.models import User
from myapp.utils import helper

Function Length

• Keep functions under 50 lines
• If longer, consider breaking into smaller functions
• Each function should do one thing well

Documentation

• Use docstrings for public functions and classes
• Follow Google or NumPy docstring style
• Include parameter types and return types

def process_user_data(user_id: str, include_history: bool = False) -> dict:
    """Process user data and return formatted result.

    Args:
        user_id: The unique identifier for the user.
        include_history: Whether to include user history. Defaults to False.

    Returns:
        A dictionary containing the processed user data.

    Raises:
        ValueError: If user_id is empty or invalid.
    """
    pass

Error Handling

Be Specific

• Catch specific exceptions, not bare except:
• Include helpful error messages

# Good
try:
    result = process_data(data)
except ValueError as e:
    logger.error(f"Invalid data format: {e}")
    raise
except ConnectionError as e:
    logger.error(f"Failed to connect: {e}")
    return None

# Bad
try:
    result = process_data(data)
except:
    pass

Security

Never Hardcode Secrets

• Use environment variables or secret managers
• Never commit credentials to version control

# Good
import os
api_key = os.environ.get("API_KEY")

# Bad
api_key = "sk-1234567890abcdef"