Token导航 LogoToken导航

Agent Skill · 研究检索

clean-code-review(干净的代码审查)

clean-code-review 用于查找、检索和筛选相关信息,适合在 Local Agent 中需要根据关键词、任务场景或来源线索快速定位候选结果时使用。可结合来源仓库、安装命令和原始 README 继续核验具体用法。安装前建议确认权限范围、维护状态,以及是否会触发联网、命令执行或文件读写。

clean-code-review 是研究检索类 Agent Skill,可用宿主为 Local Agent,通过unknown获取,权限为可写文件;许可证信息还不明确,暂无第三方扫描结果,按安装量在研究检索类的 73,675 个 Skill 里排第 68,884。

总安装 222周安装 9下载 70核验 2026-05-01
研究检索可写文件unknown未标认证来源可访问许可证需确认审计未展示Local Agent

安装说明

本站只整理中文说明,不托管安装包,也不代用户安装。
许可证
unknown

安装方式

通过对话安装

复制提示词发给支持本地命令或 Skills 的 AI 助手,先确认命令和权限,再让它执行。

请帮我安装这个 Agent Skill:clean-code-review(干净的代码审查)
来源仓库:https://skills.volces.com
仓库路径:clean-code-review
安装命令:
OpenClaw / Moltbot / Clawbot
安装前请先检查当前环境是否支持对应 CLI,并向我确认将要执行的命令、安装目录、联网范围和文件读写权限;确认后再执行。

命令行安装

复制命令到本机终端执行。该命令会通过 对应宿主 从第三方来源获取 Skill;本站只展示命令,不托管安装包,也不自动执行。

skills.sh安装方式未标明
OpenClaw / Moltbot / Clawbot

这个 Skill 是做什么的

研究检索类 Skill 主要解决「找资料、核对事实、把来源整理清楚」的问题,通常会让助手先检索再总结。

clean-code-review 用于查找、检索和筛选干净代码审查相关信息。适合在 Local Agent 中根据关键词或代码审查需求快速定位候选结果。可通过 OpenClaw / Moltbot / Clawbot 安装,具体用法请参考来源仓库。安装前建议确认权限范围、维护状态及是否触发联网或文件读写。

当前分类为研究检索,实际能力需以官方文档为准。

来源仓库内路径:clean-code-review

上手步骤

4 步
  1. 1

    确认自己的宿主工具

    公开资料标注支持 Local Agent;先确认自己用的是其中之一,再按对应工具的 Skills 目录放置。

  2. 2

    拿到安装命令

    可以直接复制的命令是 OpenClaw / Moltbot / Clawbot;命令里的占位符换成自己的路径或密钥再执行。

  3. 3

    安装后重启宿主

    命令执行完要完全退出宿主工具再打开,让新装的 Skill 被重新扫描到,否则列表里可能看不到。

  4. 4

    按最小权限试用

    它需要可写文件,先在测试目录或测试账号里跑,确认没问题再放到正式项目。

常见问题

6 条

clean-code-review 怎么安装?

在终端执行 OpenClaw / Moltbot / Clawbot。装完完全退出宿主工具再打开,让它重新扫描本地 Skill。

clean-code-review 支持哪些 AI 工具?

公开资料标注支持 Local Agent。其他工具只要能读取本地 Skill 目录,一般也能手动放进去,但要自己确认触发方式是否一致。

clean-code-review 需要什么权限?

权限级别是可写文件。该 Skill 可能写入或修改本地文件,使用前需要确认目标目录和修改范围。

clean-code-review 是免费的吗?

公开资料没有标注明确许可证,收费和商用限制都无法确认。正式使用前请到来源仓库核对授权说明。

clean-code-review 用的人多吗?

公开数据是累计安装 222 次,近一周安装 9 次,下载 70 次。这些数字来自第三方目录统计,只能当作热度参考,不代表质量或安全性。

clean-code-review 安全吗?

没有查到第三方扫描结果,安全性无法从公开资料判断,安装前请自行审查源码、依赖和命令行为。

值得选的地方和要先想清楚的地方

值得选的地方

  • 专注 Local Agent 一种宿主,触发方式和使用习惯不用来回切换。

先想清楚的地方

  • 许可证没有明确标注,能不能商用、能不能再分发都不确定。
  • 没有第三方扫描结果,安全性只能靠自己读源码判断。
  • 第三方目录整理,版本更新和行为细节以来源仓库为准。

适不适合你

更适合

  • 已经在用 Local Agent,想直接扩展它能力的人。
  • 需要处理研究检索类任务、又不想每次都手写提示词的人。

不太适合

  • 需要明确授权才能用到公司项目里的场景。
  • 要求有安全扫描结论才肯安装的团队。
  • 需要厂商级技术支持和 SLA 的生产系统。

研究检索类 Skill 对比

6 个同类
Skill总安装GitHub 星权限许可证扫描
find-skills

寻找技能 · Codex、Claude

31,199,81316,486操作浏览器MIT有提醒
self-improving-agent

自我改进 Agent · OpenClaw

16,464,0023,396接触敏感数据MIT-0通过
soultrace

灵魂追踪 · Codex、Claude

7,588,94412需要联网未明确有提醒
self-improving

自我完善 · OpenClaw

6,829,2841,046需要联网MIT-0通过
microsoft-foundry

微软代工厂 · Codex、Claude

6,614,144764调用外部服务未明确有提醒
clean-code-review

干净的代码审查 · 本页

222—可写文件未明确未收录

按安装量取同分类前几名;数字来自第三方目录统计,只作热度参考,不代表质量结论。

SKILL.md 原文

Clean Code

Be concise, direct, and solution-focused. Clean code reads like well-written prose — every name reveals intent, every function does one thing, and every abstraction earns its place.

Installation

OpenClaw / Moltbot / Clawbot

npx clawhub@latest install clean-code

Core Principles

PrincipleRulePractical Test
SRPSingle Responsibility — each function/class does ONE thing"Can I describe what this does without using 'and'?"
DRYDon't Repeat Yourself — extract duplicates, reuse"Have I written this logic before?"
KISSKeep It Simple — simplest solution that works"Is there a simpler way to achieve this?"
YAGNIYou Aren't Gonna Need It — don't build unused features"Does anyone need this right now?"
Boy ScoutLeave code cleaner than you found it"Is this file better after my change?"

Naming Rules

Names are the most important documentation. A good name eliminates the need for a comment.

ElementConventionBadGood
VariablesReveal intentn, d, tmpuserCount, elapsed, activeUsers
FunctionsVerb + nounuser(), calc()getUserById(), calculateTotal()
BooleansQuestion formactive, flagisActive, hasPermission, canEdit
ConstantsSCREAMING_SNAKEmax, timeoutMAX_RETRY_COUNT, REQUEST_TIMEOUT_MS
ClassesNoun, singularManager, DataUserRepository, OrderService
EnumsPascalCase values'pending' stringStatus.Pending
Rule: If you need a comment to explain a name, rename it.

Naming Anti-Patterns

Anti-PatternProblemFix
Cryptic abbreviations (usrMgr, cfg)Unreadable in 6 monthsSpell it out — IDE autocomplete makes long names free
Generic names (data, info, item, handler)Says nothing about purposeUse domain-specific names that reveal intent
Misleading names (getUserList returns one user)Actively deceives readersMatch name to behavior, or change the behavior
Hungarian notation (strName, nCount, IUser)Redundant with type systemLet TypeScript/IDE show types; names describe purpose

Function Rules

RuleGuidelineWhy
SmallMax 20 lines, ideally 5-10Fits in your head
One ThingDoes one thing, does it wellTestable and nameable
One LevelOne level of abstraction per functionReadable top to bottom
Few ArgsMax 3 arguments, prefer 0-2Easy to call correctly
No Side EffectsDon't mutate inputs unexpectedlyPredictable behavior

Guard Clauses

Flatten nested conditionals with early returns. Never nest deeper than 2 levels.

// BAD — 5 levels deep
function processOrder(order: Order) {
  if (order) {
    if (order.items.length > 0) {
      if (order.customer) {
        if (order.customer.isVerified) {
          return submitOrder(order);
        }
      }
    }
  }
  throw new Error('Invalid order');
}

// GOOD — guard clauses flatten the structure
function processOrder(order: Order) {
  if (!order) throw new Error('No order');
  if (!order.items.length) throw new Error('No items');
  if (!order.customer) throw new Error('No customer');
  if (!order.customer.isVerified) throw new Error('Customer not verified');

  return submitOrder(order);
}

Parameter Objects

When a function needs more than 3 arguments, use an options object.

// BAD — too many parameters, order matters
createUser('John', 'Doe', 'john@example.com', 'secret', 'admin', 'Engineering');

// GOOD — self-documenting options object
createUser({
  firstName: 'John',
  lastName: 'Doe',
  email: 'john@example.com',
  password: 'secret',
  role: 'admin',
  department: 'Engineering',
});

Code Structure Patterns

PatternWhen to ApplyBenefit
Guard ClausesEdge cases at function startFlat, readable flow
Flat > NestedAny nesting beyond 2 levelsReduced cognitive load
CompositionComplex operationsSmall, testable pieces
ColocationRelated code across filesEasier to find and change
Extract FunctionComments separating "sections"Self-documenting code

Composition Over God Functions

// BAD — god function doing everything
async function processOrder(order: Order) {
  // Validate... (15 lines)
  // Calculate totals... (15 lines)
  // Process payment... (10 lines)
  // Send notifications... (10 lines)
  // Update inventory... (10 lines)
  return { success: true };
}

// GOOD — composed of small, focused functions
async function processOrder(order: Order) {
  validateOrder(order);
  const totals = calculateOrderTotals(order);
  const payment = await processPayment(order.customer, totals);
  await sendOrderConfirmation(order, payment);
  await updateInventory(order.items);
  return { success: true, orderId: payment.orderId };
}

Return Type Consistency

Functions should return consistent types. Use discriminated unions for multiple outcomes.

// BAD — returns different types
function getUser(id: string) {
  const user = database.find(id);
  if (!user) return false;     // boolean
  if (user.isDeleted) return null; // null
  return user;                 // User
}

// GOOD — discriminated union
type GetUserResult =
  | { status: 'found'; user: User }
  | { status: 'not_found' }
  | { status: 'deleted' };

function getUser(id: string): GetUserResult {
  const user = database.find(id);
  if (!user) return { status: 'not_found' };
  if (user.isDeleted) return { status: 'deleted' };
  return { status: 'found', user };
}

Anti-Patterns

Anti-PatternProblemFix
Comment every lineNoise obscures signalDelete obvious comments; comment *why*, not *what*
Helper for one-linerUnnecessary indirectionInline the code
Factory for 2 objectsOver-engineeringDirect instantiation
utils.ts with 1 functionJunk drawer filePut code where it's used
Deep nestingUnreadable flowGuard clauses and early returns
Magic numbersUnclear intentNamed constants
God functionsUntestable, unreadableSplit by responsibility
Commented-out codeDead code confusionDelete it; git remembers
TODO sprawlNever gets doneTrack in issue tracker, not code
Premature abstractionWrong abstraction is worse than noneWait for 3+ duplicates before abstracting
Copy-paste programmingDuplicated bugsExtract shared logic
Exception-driven control flowSlow and confusingUse explicit conditionals
Stringly-typed codeTypos and missed casesUse enums or union types
Callback hellPyramid of doomUse async/await

Pre-Edit Safety Check

Before changing any file, answer these questions to avoid cascading breakage:

QuestionWhy
What imports this file?Dependents might break on interface changes
What does this file import?You might need to update the contract
What tests cover this?Tests might fail — update them alongside code
Is this a shared component?Multiple consumers means wider blast radius
File to edit: UserService.ts
├── Who imports this? → UserController.ts, AuthController.ts
├── Do they need changes too? → Check function signatures
└── What tests cover this? → UserService.test.ts
Rule: Edit the file + all dependent files in the SAME task. Never leave broken imports or missing updates.

Self-Check Before Completing

Before marking any task complete, verify:

CheckQuestion
Goal met?Did I do exactly what was asked?
Files edited?Did I modify all necessary files, including dependents?
Code works?Did I verify the change compiles and runs?
No errors?Do lint and type checks pass?
Nothing forgotten?Any edge cases or dependent files missed?

NEVER Do

  1. NEVER add comments that restate the code — if the code needs a comment to explain *what* it does, rename things until it doesn't
  2. NEVER create abstractions for fewer than 3 use cases — premature abstraction is worse than duplication
  3. NEVER leave commented-out code in the codebase — delete it; version control exists for history
  4. NEVER write functions longer than 20 lines — extract sub-functions until each does one thing
  5. NEVER nest deeper than 2 levels — use guard clauses, early returns, or extract functions
  6. NEVER use magic numbers or strings — define named constants with clear semantics
  7. NEVER edit a file without checking what depends on it — broken imports and missing updates are the most common source of bugs in multi-file changes
  8. NEVER leave a task with failing lint or type checks — fix all errors before marking complete

References

Detailed guides for specific clean code topics:

ReferenceDescription
Anti-Patterns21 common mistakes with bad/good code examples across naming, functions, structure, and comments
Code SmellsClassic code smells catalog with detection patterns — Bloaters, OO Abusers, Change Preventers, Dispensables, Couplers
Refactoring CatalogEssential refactoring patterns with before/after examples and step-by-step mechanics

平台分布

Local Agent

87%
按下载量换算61

安全审计

暂无安全审计结果可展示。

权限和风险

可写文件

该 Skill 可能写入或修改本地文件,使用前需要确认目标目录和修改范围。

安装前确认

本站仅展示第三方公开信息,不托管安装包,不提供自动安装或运行环境。安装前应自行审查源码、依赖和命令行为。目前只有一个公开来源,建议同时核对源仓库。

继续浏览同类 Skills