code-prose

Rewrites code identifiers and comments using Orwell's plain-language rules without changing behavior.

4.8k|376|Updated Mar 16, 2023
One-click install
npx skills add https://github.com/EpicenterHQ/epicenter --skill code-prose
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: code-prose
Source: https://github.com/EpicenterHQ/epicenter/tree/main/.agents/skills/code-prose
Command: npx skills add https://github.com/EpicenterHQ/epicenter --skill code-prose

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Codebases accumulate vague names, redundant compound identifiers, and comments that narrate history instead of explaining constraints, making code hard to read for anyone without context on how it was written.

Core Features & Use Cases

  • Name Review: Applies Orwell's rules to variable and function names, enforcing one word per concept, cutting redundant words, and replacing Latinate vocabulary with short Anglo-Saxon alternatives.
  • Comment Editing: Keeps comments that explain non-obvious constraints and deletes comments that restate self-evident code or narrate change history.
  • Overfitting Detection: Rewrites names and comments that only make sense to someone who watched the code being written, so the code stands on its own.
  • Use Case: After finishing a feature branch, run this Skill to give identifiers and comments an editorial pass before opening a pull request, then run the project's existing checks to confirm behavior is unchanged.

Quick Start

Review the names and comments in my current branch and rewrite them using plain-language rules without changing any behavior.

Frequently Asked Questions about code-prose

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I improve variable and function names in my code?

Apply plain-language rules: use one word per concept, cut words the surrounding context already carries, and prefer short Anglo-Saxon words over Latinate ones. This Skill reviews names in your branch and rewrites them without changing behavior.

When should I delete a code comment?

Delete comments that narrate change history from a conversation or restate code whose behavior is self-evident. Keep comments that explain non-obvious constraints, complex implementations, or side effects the code cannot show.

Does renaming identifiers change how my code behaves?

No. This Skill only touches names and comments within the scope you specify and instructs running the project's existing checks afterward to confirm behavior is unchanged.

What is overfitting in code comments and names?

Overfitting means a name or comment only makes sense to someone who watched the code being written, such as references to a conversation or PR. The fix is rewriting it against the codebase's own vocabulary so a new reader understands it.

When should I not use this code-prose review approach?

Avoid it when you need structural changes like file organization, derivable state, or compatibility refactors. The Skill explicitly defers that structural work to the companion refactoring skill.