mcpbeat

Code Quality Principles

athola/code-quality-principles

Applies KISS, YAGNI, and SOLID principles for clean code with reduced complexity. Use when refactoring or reviewing code for over-engineering.

2k tokens
context cost
the whole folder, loaded on every use
1
files
instructions only
0
copies elsewhere
how many repositories repackaged it
324
stars on the repo
on the repository, not the skill itself

Install

one command, takes just this skill from the repository
npx skills add https://github.com/athola/claude-night-market --skill code-quality-principles

The instruction itself

22 sections, as written by the author

Code Quality Principles

Guidance on KISS, YAGNI, and SOLID principles with language-specific examples.

When To Use

  • Improving code readability and maintainability
  • Applying SOLID, KISS, YAGNI principles during refactoring

When NOT To Use

  • Throwaway scripts or one-time data migrations
  • Performance-critical code where readability trades are justified

KISS (Keep It Simple, Stupid)

Principle: Avoid unnecessary complexity. Prefer obvious solutions over clever ones.

Guidelines

| Prefer | Avoid |

|--------|-------|

| Simple conditionals | Complex regex for simple checks |

| Explicit code | Magic numbers/strings |

| Standard patterns | Clever shortcuts |

| Direct solutions | Over-abstracted layers |

Python Example

# Bad: Overly clever one-liner
users = [u for u in (db.get(id) for id in ids) if u and u.active and not u.banned]

# Good: Clear and readable
users = []
for user_id in ids:
    user = db.get(user_id)
    if user and user.active and not user.banned:
        users.append(user)

Rust Example

// Bad: Unnecessary complexity
fn process(data: &[u8]) -> Result<Vec<u8>, Box<dyn std::error::Error>> {
    data.iter()
        .map(|&b| b.checked_add(1).ok_or("overflow"))
        .collect::<Result<Vec<_>, _>>()
        .map_err(|e| e.into())
}

// Good: Simple and clear
fn process(data: &[u8]) -> Result<Vec<u8>, &'static str> {
    let mut result = Vec::with_capacity(data.len());
    for &byte in data {
        result.push(byte.checked_add(1).ok_or("overflow")?);
    }
    Ok(result)
}

YAGNI (You Aren't Gonna Need It)

Principle: Don't implement features until they are actually needed.

Guidelines

| Do | Don't |

|----|-------|

| Solve current problem | Build for hypothetical futures |

| Add when 3rd use case appears | Create abstractions for 1 use case |

| Delete dead code | Keep "just in case" code |

| Minimal viable solution | Premature optimization |

Python Example

# Bad: Premature abstraction for one use case
class AbstractDataProcessor:
    def process(self, data): ...
    def validate(self, data): ...
    def transform(self, data): ...

class CSVProcessor(AbstractDataProcessor):
    def process(self, data):
        return self.transform(self.validate(data))

# Good: Simple function until more cases appear
def process_csv(data: list[str]) -> list[dict]:
    return [parse_row(row) for row in data if row.strip()]

TypeScript Example

// Bad: Over-engineered config system
interface ConfigProvider<T> {
  get<K extends keyof T>(key: K): T[K];
  set<K extends keyof T>(key: K, value: T[K]): void;
  watch<K extends keyof T>(key: K, callback: (v: T[K]) => void): void;
}

// Good: Simple config for current needs
const config = {
  apiUrl: process.env.API_URL || 'http://localhost:3000',
  timeout: 5000,
};

SOLID Principles

Single Responsibility Principle

Each module/class should have one reason to change.

# Bad: Multiple responsibilities
class UserManager:
    def create_user(self, data): ...
    def send_welcome_email(self, user): ...  # Email responsibility
    def generate_report(self, users): ...     # Reporting responsibility

# Good: Separated responsibilities
class UserRepository:
    def create(self, data): ...

class EmailService:
    def send_welcome(self, user): ...

class UserReportGenerator:
    def generate(self, users): ...

Open/Closed Principle

Open for extension, closed for modification.

# Bad: Requires modification for new types
def calculate_area(shape):
    if shape.type == "circle":
        return 3.14 * shape.radius ** 2
    elif shape.type == "rectangle":
        return shape.width * shape.height
    # Must modify to add new shapes

# Good: Extensible without modification
from abc import ABC, abstractmethod

class Shape(ABC):
    @abstractmethod
    def area(self) -> float: ...

class Circle(Shape):
    def __init__(self, radius: float):
        self.radius = radius
    def area(self) -> float:
        return 3.14 * self.radius ** 2

Liskov Substitution Principle

Subtypes must be substitutable for their base types.

# Bad: Violates LSP - Square changes Rectangle behavior
class Rectangle:
    def set_width(self, w): self.width = w
    def set_height(self, h): self.height = h

class Square(Rectangle):  # Breaks when used as Rectangle
    def set_width(self, w):
        self.width = self.height = w  # Unexpected side effect

# Good: Separate types with common interface
class Shape(ABC):
    @abstractmethod
    def area(self) -> float: ...

class Rectangle(Shape):
    def __init__(self, width: float, height: float): ...

class Square(Shape):
    def __init__(self, side: float): ...

Interface Segregation Principle

Clients shouldn't depend on interfaces they don't use.

// Bad: Fat interface
interface Worker {
  work(): void;
  eat(): void;
  sleep(): void;
}

// Good: Segregated interfaces
interface Workable {
  work(): void;
}

interface Feedable {
  eat(): void;
}

// Clients only implement what they need
class Robot implements Workable {
  work(): void { /* ... */ }
}

Dependency Inversion Principle

Depend on abstractions, not concretions.

# Bad: Direct dependency on concrete class
class OrderService:
    def __init__(self):
        self.db = PostgresDatabase()  # Tight coupling

# Good: Depend on abstraction
from abc import ABC, abstractmethod

class Database(ABC):
    @abstractmethod
    def save(self, data): ...

class OrderService:
    def __init__(self, db: Database):
        self.db = db  # Injected abstraction

Quick Reference

| Principle | Question to Ask | Red Flag |

|-----------|-----------------|----------|

| KISS | "Is there a simpler way?" | Complex solution for simple problem |

| YAGNI | "Do I need this right now?" | Building for hypothetical use cases |

| SRP | "What's the one reason to change?" | Class doing multiple jobs |

| OCP | "Can I extend without modifying?" | Switch statements for types |

| LSP | "Can subtypes replace base types?" | Overridden methods with side effects |

| ISP | "Does client need all methods?" | Empty method implementations |

| DIP | "Am I depending on abstractions?" | new keyword in business logic |

When Principles Conflict

  • KISS vs SOLID: For small projects, KISS wins. Add SOLID patterns as complexity grows.
  • YAGNI vs DIP: Don't add abstractions until you have 2+ implementations.
  • Readability vs DRY: Prefer slight duplication over wrong abstraction.

Integration with Code Review

When reviewing code, check:

  • [ ] No unnecessary complexity (KISS)
  • [ ] No speculative features (YAGNI)
  • [ ] Each class has single responsibility (SRP)
  • [ ] No god classes (> 500 lines)
  • [ ] Dependencies are injected, not created (DIP)

Verification: Run wc -l <file> to check line counts and rg -c "class " <file> (or grep -c "class " <file>) to count classes per file.

  • imbue:karpathy-principles - The "Simplicity First" principle wraps KISS, YAGNI, and SOLID into a four-principle synthesis derived from Karpathy's observations on LLM coding pitfalls
  • See docs/quality-gates.md#skill-level-quality-gate-composition for the full gate-skill federation graph

Exit Criteria

  • [ ] Every proposed code change checked against the integration

review checklist: no unnecessary complexity (KISS), no speculative

features (YAGNI), single responsibility per class (SRP), no god

classes over 500 lines, dependencies injected not created (DIP)

  • [ ] When KISS and SOLID conflict, the resolution is documented:

KISS wins for small projects, SOLID patterns applied as complexity

grows. The choice is explicit, not silent

  • [ ] wc -l <file> run on any modified file and result noted if

the file exceeds 500 lines (god-class threshold)

  • [ ] No new abstraction introduced with only one implementor unless

it serves as a mock boundary for testing

How to use it

Copy the folder

Take athola/code-quality-principles from the repository into ~/.claude/skills for personal use, or into .claude/skills inside a project.

Check the name does not clash

The agent identifies a skill by the name field in its header. Two skills with the same name cannot sit side by side — one of them will be ignored.