How do I install the Clean Code skill?

Run npx skillmds add majiayu000/clean-code-2 in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.

Is Clean Code safe to use?

This skill has not completed SkillMD's automated safety review yet. Capability flags: docs only. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.

Which AI agents work with Clean Code?

This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.

Is Clean Code free to use?

Yes. Installing skills from SkillMD is free, and the skill stays under its author's original license.

Who published Clean Code?

majiayu000 (@majiayu000) published this skill. Their other Agent Skills are listed on their SkillMD profile.

Clean Code by majiayu000 — Agent Skill

Name: Clean Code
Rating: 4.2 (5 reviews)
Author: majiayu000

File contents

Auto mode (skip confirmations):

/clean-code -a dashboard

With save (output to .claude/output/clean-code/):

/clean-code -s -a refactor api

What it does:

Scans codebase for issues (React, Next.js, Zustand, TanStack Query)
Loads relevant best practices docs
Applies clean code improvements
Verifies with build and tests
Commits changes

Flag	Description
`-a`	Auto mode: skip confirmations
`-e`	Economy mode: no subagents, direct tools only
`-s`	Save mode: output to `.claude/output/clean-code/`
`-r`	Resume mode: continue from previous task
`--react`	Force load React 19 patterns
`--nextjs`	Force load Next.js 15/16 patterns
`--zustand`	Force load Zustand v5 patterns
`--query`	Force load TanStack Query v5 patterns

Variable	Type	Description
`{task_description}`	string	What to analyze
`{task_id}`	string	Kebab-case identifier
`{auto_mode}`	boolean	Skip confirmations
`{economy_mode}`	boolean	No subagents
`{save_mode}`	boolean	Save outputs
`{detected_tech}`	object	Technologies found
`{issues}`	array	Issues found

File	When Loaded
`references/general-clean-code.md`	Always
`references/react-clean-code.md`	React detected / `--react`
`references/nextjs-clean-code.md`	Next.js detected / `--nextjs`
`references/zustand-best-practices.md`	Zustand detected / `--zustand`
`references/tanstack-query-best-practices.md`	No data fetching / `--query`

Load steps/step-01-scan.md

Step	File	Purpose
01	`step-01-scan.md`	Init + analyze
02	`step-02-apply.md`	Load docs + apply
03	`step-03-verify.md`	Build + commit

---name: clean-code-2description: Comprehensive clean code workflow that analyzes codebase, recommends best practices, and applies clean code principles for React, Next.js, and modern tooling.--- <objective> Fast, systematic clean code improvements using parallel agents. </objective> <quick_start> **Basic usage (analyze and apply clean code principles):** ```bash /clean-code auth feature ``` **Auto mode (skip confirmations):** ```bash /clean-code -a dashboard ``` **With save (output to `.claude/output/clean-code/`):** ```bash /clean-code -s -a refactor api ``` **What it does:** 1. Scans codebase for issues (React, Next.js, Zustand, TanStack Query) 2. Loads relevant best practices docs 3. Applies clean code improvements 4. Verifies with build and tests 5. Commits changes </quick_start> <parameters> | Flag | Description | |------|-------------| | `-a` | Auto mode: skip confirmations | | `-e` | Economy mode: no subagents, direct tools only | | `-s` | Save mode: output to `.claude/output/clean-code/` | | `-r` | Resume mode: continue from previous task | | `--react` | Force load React 19 patterns | | `--nextjs` | Force load Next.js 15/16 patterns | | `--zustand` | Force load Zustand v5 patterns | | `--query` | Force load TanStack Query v5 patterns | <examples> ```bash /clean-code auth feature # Basic /clean-code -a dashboard # Auto mode /clean-code -e -a fix types # Economy + auto /clean-code -s refactor api # Save outputs /clean-code -r auth-feature # Resume /clean-code --nextjs --query data # Force docs ``` </examples> </parameters> <workflow> ``` SCAN → APPLY → VERIFY │ │ │ │ │ └─ Build, test, commit │ └─ Load docs, recommend, apply └─ Parse flags, detect tech, find issues ``` </workflow> <state_variables> | Variable | Type | Description | |----------|------|-------------| | `{task_description}` | string | What to analyze | | `{task_id}` | string | Kebab-case identifier | | `{auto_mode}` | boolean | Skip confirmations | | `{economy_mode}` | boolean | No subagents | | `{save_mode}` | boolean | Save outputs | | `{detected_tech}` | object | Technologies found | | `{issues}` | array | Issues found | </state_variables> <reference_files> | File | When Loaded | |------|-------------| | `references/general-clean-code.md` | Always | | `references/react-clean-code.md` | React detected / `--react` | | `references/nextjs-clean-code.md` | Next.js detected / `--nextjs` | | `references/zustand-best-practices.md` | Zustand detected / `--zustand` | | `references/tanstack-query-best-practices.md` | No data fetching / `--query` | </reference_files> <entry_point> Load `steps/step-01-scan.md` </entry_point> <step_files> | Step | File | Purpose | |------|------|---------| | 01 | `step-01-scan.md` | Init + analyze | | 02 | `step-02-apply.md` | Load docs + apply | | 03 | `step-03-verify.md` | Build + commit | </step_files> <execution_rules> - Load one step at a time - Use parallel agents in step-01 (unless economy mode) - Always run build before completing - Follow patterns from reference files exactly </execution_rules> <success_criteria> - Technologies correctly detected (React, Next.js, Zustand, TanStack Query) - Issues identified and documented - Relevant reference docs loaded - Clean code improvements applied - Build passes without errors - Tests pass (if tests exist) - Changes committed with clear message - No hacks or shortcuts used </success_criteria>

Frequently asked questions