When is code duplication better than abstraction?

Duplication is preferred when requirements are still evolving or when different use cases share a similar structure but require different behaviors. Duplication is significantly cheaper to fix than unwinding a wrong or coupled abstraction.

What is the 'Rule of Three' used by this skill?

The Rule of Three suggests that code should not be abstracted until it appears in at least three places. This ensures the pattern is genuine and provides enough context to create a stable abstraction rather than a premature one.

What are the hidden costs of abstraction mentioned?

The primary hidden costs include accidental coupling between unrelated modules, layers of indirection that obscure bugs, and organizational inertia that makes future refactoring more difficult.

How does this skill help fix 'incomprehensible spaghetti' code?

It guides the user to inline the abstraction back into each caller, delete unnecessary logic for each specific use case, and then re-extract a proper abstraction only once the patterns have emerged naturally.

Code Architecture & Abstraction Guide

Name: Code Architecture & Abstraction Guide
Author: Microck

byMicrock

•

Developer Tools

Guides architectural decisions by identifying when to abstract logic versus maintaining intentional duplication to prevent technical debt.

This skill provides Claude with expert-level heuristics for software architecture, focusing on the Avoid Hasty Abstractions (AHA) philosophy. It helps developers navigate the tension between DRY (Don't Repeat Yourself) and WET (Write Everything Twice) by applying the 'Rule of Three' and identifying the 'Wrong Abstraction' pattern. By providing actionable strategies for inlining over-engineered code and evaluating the hidden costs of facades, this skill ensures codebases remain flexible, maintainable, and easy to refactor as requirements evolve.

Key Features

01Distinguishes between structural similarity and genuine behavioral identity

02Identifies 'Wrong Abstraction' patterns in bloated, conditional-heavy functions

03Evaluates Facade pattern trade-offs for UI components and complex logic

042 GitHub stars

05Applies the 'Rule of Three' to determine pattern stability before abstraction

06Provides step-by-step recovery paths for inlining and fixing leaky abstractions

Use Cases

01Deciding whether to create a shared utility or keep code duplicated across two modules

02Reviewing architectural designs to prevent accidental coupling between unrelated features

03Refactoring a complex React component that has become difficult to extend with new props

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add microck/opendots-microck flpbalada-code-architecture-wrong-abstraction

For use in Claude.ai and ChatGPT

Key Features

01Distinguishes between structural similarity and genuine behavioral identity

02Identifies 'Wrong Abstraction' patterns in bloated, conditional-heavy functions

03Evaluates Facade pattern trade-offs for UI components and complex logic

042 GitHub stars

05Applies the 'Rule of Three' to determine pattern stability before abstraction

06Provides step-by-step recovery paths for inlining and fixing leaky abstractions

Use Cases

01Deciding whether to create a shared utility or keep code duplicated across two modules

02Reviewing architectural designs to prevent accidental coupling between unrelated features

03Refactoring a complex React component that has become difficult to extend with new props

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add microck/opendots-microck flpbalada-code-architecture-wrong-abstraction

For use in Claude.ai and ChatGPT