---
title: Clean Code Skills for Claude Code
canonical: https://skillselion.com/guides/clean-code-skills-claude-code
content_type: text/markdown
description: "The most-installed Clean Code skill in the Skillselion catalog is sickn33's clean-code (13,115 installs), which applies Robert C."
---

# Clean Code Skills for Claude Code: Uncle Bob Rules and Code Review

Title: Clean Code Skills for Claude Code

> The most-installed Clean Code skill in the Skillselion catalog is sickn33's clean-code (13,115 installs), which applies Robert C.

Canonical: https://skillselion.com/guides/clean-code-skills-claude-code

The most-installed Clean Code skill in the Skillselion catalog is sickn33's clean-code (13,115 installs), which applies Robert C. Martin's rules for names, functions, comments and error handling. Add clean-code-guard (4,961 installs) to check every diff before commit, and wondelai's smell catalog (7,377 installs) for refactoring.

## Key facts

- Last updated: November 10, 2026

## Article

Several skills now load Robert C. Martin's *Clean Code* into Claude Code. The most-installed in our catalog is the [clean-code skill from sickn33's agentic-awesome-skills](/skills/sickn33/agentic-awesome-skills/clean-code) (13,115 installs, 50,109 GitHub stars), and the right pick depends on whether you want the agent to write clean code or to check it.

## Key takeaways

- sickn33's clean-code (13,115 installs) condenses Uncle Bob's book into rules for names, functions, comments, formatting, objects, error handling and tests.
- wondelai's [clean-code smell catalog](/skills/wondelai/skills/clean-code) (7,377 installs) is based on Chapter 17 of the book and lists smells by category with a refactoring for each.
- amelnagdy's [clean-code-guard](/skills/amelnagdy/guard-skills/clean-code-guard) (4,961 installs) runs after the agent writes code and checks Clean Code, SOLID, DRY, KISS, YAGNI and LLM failure modes before commit.
- davila7's clean-code (1,142 installs) is a short, strict style card: functions of 20 lines max, at most 3 arguments, nesting at most 2 levels deep.
- The [testing and review category](/category/testing-review) lists the wider set of review skills.

## What does a Clean Code skill change in Claude Code?

It gives the agent a fixed standard to write and review against instead of its own habits. sickn33's skill (13,115 installs, skills.sh registry) turns the book into direct rules: intention-revealing names like `elapsedTimeInDays` instead of `d`, functions that do one thing at one level of abstraction, 0 arguments as the ideal and 3 or more only with strong justification, exceptions instead of return codes, and no returning or passing `null`. It also covers the Law of Demeter, so chains like `a.getB().getC().doSomething()` get flagged. The rules apply when writing new code, reviewing pull requests and refactoring legacy code.

## What is the best Clean Code skill for Claude Code?

Start with sickn33's clean-code, the most-installed Uncle Bob skill in our catalog at 13,115 installs, in a repo with 50,109 GitHub stars (skills.sh registry, GitHub). Install it with:

`npx skills add https://github.com/sickn33/agentic-awesome-skills --skill clean-code`

![sickn33 avatar on GitHub](https://github.com/sickn33.png?size=200)

If you want depth over breadth, wondelai's clean-code (7,377 installs, 2,759 GitHub stars) is a smell catalog with IDs such as C1 to C5 for comment smells and E1 and E2 for build and test environment smells, each with a bad example and a fix. It is more useful during refactoring than during first drafts. Martin's own [Clean Coder blog](https://blog.cleancoder.com/) is the place to check his intent when a rule seems too strict.

## How do you make Claude Code review its own code?

Add a guard skill that runs after the first implementation pass. clean-code-guard (4,961 installs, 1,320 GitHub stars) has three modes: a guard pass that checks the diff before commit, a live mode you invoke before a risky edit, and a review mode that walks a checklist and writes a findings report without editing files. It needs no MCP server, API key or script, and its references cover naming, comments, SOLID, DRY and KISS in separate files. It also says what it is not for: test code, CI config and pure architecture talk. For AI-written code that works but feels bloated, [ai-slop-cleaner from oh-my-claudecode](/skills/yeachan-heo/oh-my-claudecode/ai-slop-cleaner) (1,743 installs) runs a deletion-first cleanup with regression tests locked first.

## Which Clean Code skill covers SOLID principles?

asyrafhussin's [clean-code-principles](/skills/asyrafhussin/agent-skills/clean-code-principles) (1,966 installs) is the most structured option for SOLID. It ships 23 rules: 10 for SOLID, 12 core rules for DRY, KISS, YAGNI and related ideas, and 1 for the repository pattern, each with bad and good examples. You can point the agent at a rule by ID, such as `solid-srp-class`. clean-code-guard also has a SOLID reference file. For module boundaries rather than class rules, Matt Pocock's [codebase-design skill](/skills/mattpocock/skills/codebase-design) covers the deep-module approach (982,730 installs).

## A complete starter stack

- clean-code from sickn33 (13,115 installs): the core Uncle Bob rules while writing.
- clean-code-guard (4,961 installs): a check on every diff before commit.
- clean-code smell catalog from wondelai (7,377 installs): for refactoring sessions.
- clean-code-principles (1,966 installs): when the team reviews against SOLID by rule ID.

## Common pitfalls

- Loading two rule sets with different numbers. davila7's skill caps functions at 20 lines while sickn33's only says "small", so pick one standard per repo.
- Running the guard skill in the same pass that writes the code. clean-code-guard is meant for a second pass; ai-slop-cleaner even splits writer and reviewer into separate steps.
- Refactoring without tests. sickn33's skill quotes the Three Laws of TDD, and ai-slop-cleaner locks behaviour with regression tests before it deletes a line.

Pick one writing skill and one guard, then compare more on the [best testing and review skills](/best/skills-for-testing-review) page.

## Related

- [/best/skills-for-testing-review](https://skillselion.com/best/skills-for-testing-review)
- [/category/testing-review](https://skillselion.com/category/testing-review)
- [/guides/best-code-review-skills-claude-code](https://skillselion.com/guides/best-code-review-skills-claude-code)
- [/guides/karpathy-guidelines-skill-claude-code](https://skillselion.com/guides/karpathy-guidelines-skill-claude-code)

## FAQ

### What is the best Clean Code skill for Claude Code?

sickn33's clean-code is the most-installed at 13,115 installs and covers names, functions, comments, formatting, error handling and tests. See [clean-code](/skills/sickn33/agentic-awesome-skills/clean-code).

### Is there an Uncle Bob Clean Code skill for Claude?

Yes. sickn33's clean-code (13,115 installs) is built on Robert C. Martin's book, and wondelai's clean-code (7,377 installs) follows its Chapter 17 smell list. See [wondelai's clean-code](/skills/wondelai/skills/clean-code).

### How do I install a Clean Code skill?

Run `npx skills add https://github.com/sickn33/agentic-awesome-skills --skill clean-code` in your project. sickn33's clean-code has 13,115 installs, the most of any Clean Code skill in our catalog. More options are on the [best testing and review skills](/best/skills-for-testing-review) page.

### Which skill checks AI-generated code for Clean Code violations?

clean-code-guard (4,961 installs) reviews generated code for Clean Code, SOLID, DRY, KISS, YAGNI and LLM failure modes before commit. See [clean-code-guard](/skills/amelnagdy/guard-skills/clean-code-guard).

### Is there a Clean Code skill for SOLID principles?

Yes. asyrafhussin's clean-code-principles (1,966 installs) has 23 rules, 10 of them for SOLID, each with bad and good examples. See [clean-code-principles](/skills/asyrafhussin/agent-skills/clean-code-principles).

## Machine access

- [Sitemap](https://skillselion.com/sitemap.xml)
- [llms.txt](https://skillselion.com/llms.txt)
- [llms-full.txt](https://skillselion.com/llms-full.txt)
- [Developer resources](https://skillselion.com/developers)
- [OpenAPI spec](https://skillselion.com/openapi.json)
- [MCP handshake](https://skillselion.com/.well-known/mcp)
- [API catalog](https://skillselion.com/.well-known/api-catalog)
- [AI catalog](https://skillselion.com/.well-known/ai-catalog.json)
- [Pricing](https://skillselion.com/pricing.md)
- [Auth](https://skillselion.com/auth.md)
- [AGENTS.md](https://skillselion.com/AGENTS.md)
- [Contact](https://skillselion.com/contact)
- [Home](https://skillselion.com/)
