独立运营 · 不出售排名位置  · 我们怎么评估工具 →
Anclum 提交收录
Z

z-astral/PCC

PCC (Pretty and Clean Codes) — Principal OOP AI Engineering Skill

装之前:没有可直接复制的命令 · 许可证 MIT · 链接可访问

第一步:npx pretty-and-clean-codes setup antigravity

在 GitHub 查看 ↗链接已于 2026-11-12 核查 · HTTP 200 · 打开的是厂商自己的官网
安装与配置
客户端配置(取自该仓库的安装命令)
{
  "mcpServers": {
    "z-astral-pcc": {
      "command": "npx",
      "args": ["pretty-and-clean-codes"]
    }
  }
}

解析自我们于 2026-11-05 抓取的该仓库 README;我们未执行过。

command/args 来自仓库写明的安装命令;不确定时请以该项目 README 为准。
VS Code(.vscode/mcp.json —— 键名是 servers)
{
  "servers": {
    "z-astral-pcc": {
      "command": "npx",
      "args": ["pretty-and-clean-codes"]
    }
  }
}
VS Code 顶层键是 <code>servers</code>,其余与上面那份相同。
Claude Code(命令行)
claude mcp add --transport stdio z-astral-pcc -- npx pretty-and-clean-codes
Claude Code 不收 JSON 文件,用这条命令直接加。
配置项的键名以你的客户端为准:客户端与 CLI
★ 数取值
1 ★ · z-astral/PCC · 取自 GitHub 搜索 API · (2026-11-13 20:00:50 核对)
来源明细
从该仓库的 GitHub 页面直接读取 · 条目类型:开源项目
收录与活跃
收录于 2026-11-06 · 最近见到 2026-11-14 · 共见到 9 次 · 被 1 个来源列出
评分构成
综合分 103 · 1 ★ · 1 个来源 · repo
安装
npx pretty-and-clean-codes
常见问题(答案都来自我们的记录)
怎么安装?
该仓库给出的安装命令是 npx pretty-and-clean-codes。我们的记录里它是仓库自己提供的;不保证在你的客户端上直接可用。
它适合谁 / 用来做什么?
我们把它归在这些标签下: 开源项目。标签来自条目自身,不是我们强加的分类。
谁维护它、还在更新吗?
仓库 z-astral/PCC;我们收录于 2026-11-06;最近见到 2026-11-14。这里没有"官方认证"或"质量评级"——我们只记录来源与时间。

# Pretty and Clean Codes (PCC)

以下摘录自该仓库的 README(片段) · 在 GitHub 读完整 README →

### Principal OOP AI Engineering Skill
High-Density Object-Oriented Programming • Zero Inline Noise • Terse Scope-Proportional Naming • Flat Control Flow
🇬🇧 English •
🇮🇩 Bahasa Indonesia •
🇯🇵 日本語 •
🇨🇳 简体中文 •
🇮🇳 हिन्दी •
🇷🇺 Русский
# 🇬🇧 English Documentation
> [!NOTE]
> What is Pretty and Clean Codes (PCC) in simple terms?
> Think of PCC as a "Senior Engineer Brain Chip" (Instruction Manual) for your AI coding assistant (such as Cursor, Claude Code, Google Antigravity, OpenCode, Devin, Windsurf, or GitHub Copilot). Once installed, your AI stops writing messy, overly long beginner code and starts writing clean, compact, Object-Oriented code like a Principal Software Architect.
## 💡 Why PCC? (Explained for Everyone)
If you have ever asked an AI to write code, you probably noticed three annoying habits:
1. Narrator Comments Everywhere: The AI writes useless comments on almost every line (// initialize variable, // loop through users), making your file look like a beginner's textbook.
2. Essay-Length Variable Names: The AI creates 40-character variable names like temporaryCustomerAuthenticationTokenString instead of clean, industry-standard names like tok or cfg.
3. Spaghetti if/else Pyramids: The AI nests if inside else inside if up to 6 levels deep, scattering loose functions everywhere instead of organizing them into neat Object-Oriented Programming (OOP) classes.
> [!TIP]
> Save 35%–45% of Your AI Token Quota & Speed Up Generation!
> Because PCC eliminates useless inline comments, compresses verbose variable names, and flattens nested blocks, your AI generates nearly half as many output tokens for the exact same functionality. Responses finish almost 2x faster and cost significantly less on paid API tiers.
## 🏛️ The 4 Engineering Pillars
text
┌─────────────────────────────────────────────────────────────────────────┐
│ PRETTY AND CLEAN CODES (PCC) │
├──────────────────┬──────────────────┬──────────────────┬────────────────┤
│ PILLAR I │ PILLAR II │ PILLAR III │ PILLAR IV │
│ Strict OOP │ Zero Inline │ Terse Scope │ Vertical │
│ Encapsulation │ Comments │ Naming │ Compression │
├──────────────────┼──────────────────┼──────────────────┼────────────────┤
│ • Cohesive class │ • 0 // or # in │ • Micro: it, v, │ • Guard clause │
│ architecture │ method bodies │ k, i, acc, fn │ early return │
│ • readonly state │ • 100% signature │ • Local: ctx, │ • No else │
│ • private helper │ Docblocks on │ cfg, req, res, │ after return │
│ isolation │ classes/methods│ buf, tx, err │ • 3-15 LOC max │
└──────────────────┴──────────────────┴──────────────────┴────────────────┘
### 1. Strict Object-Oriented Encapsulation (OOP)
No loose, floating functions. All business logic, state, and helpers are encapsulated inside cohesive classes (AuthSvc, TokenGuard, OrderCalc, PacketBuf) with explicit visibility modifiers (private, protected, public, readonly).
### 2. Zero Inline Comments & 100% Signature Docblocks
Zero (0) comments are allowed inside a function's body. Instead, PCC forces the AI to write a comprehensive, professional Docblock (JSDoc, TSDoc, Javadoc, or Docstring) directly above the class and method signature—documenting what the function does, its @param inputs, and its @returns output.
收录徽章(可选)
想把收录标记放进 README 就复制这行,可选,不影响收录。
[![Listed on Anclum](https://anclum.com/badge/listed.svg)](https://anclum.com/zh/s/z-astral-pcc/)
来源
github · https://github.com/z-astral/PCC
按领域浏览
开源项目
Traction
1 ★
收录于
今天

同类推荐

全部 1,671 →

所属领域与标签

每个标签都能点进去看同类的全部条目。数字是当前库里的真实条数。

浏览全部领域 →

← 返回索引