Co-Pilot / 辅助式
更新于 a month ago

claude-codepro

Mmaxritter
0.4k
maxritter/claude-codepro
80
Agent 评分

💡 摘要

Claude CodePro 是一个开发环境,强制执行测试驱动开发(TDD)并自动化代码项目的质量检查。

🎯 适合人群

寻找自动化测试解决方案的软件开发人员采用测试驱动开发(TDD)的团队监督代码质量的项目经理需要高效编码工作流程的自由职业者教授软件开发最佳实践的教育工作者

🤖 AI 吐槽:看起来很能打,但别让配置把人劝退。

安全分析中风险

风险:Medium。建议检查:是否执行 shell/命令行指令;是否发起外网请求(SSRF/数据外发);API Key/Token 的获取、存储与泄露风险。以最小权限运行,并在生产环境启用前审计代码与依赖。

Production-Grade Development Environment for Claude Code

Quality automated. Context optimized. Testing enforced. Ship with confidence. A shortcut to state-of-the-art Claude Code development, continuously improved.

Version Stars Last Commit PRs Welcome

Star this repository · 🌐 Visit the website · 🔔 Follow for updates


⚡ TL;DR

curl -fsSL https://raw.githubusercontent.com/maxritter/claude-codepro/v5.3.1/install.sh | bash

Then run ccp and /sync once to sync rules and skills. Use /spec or the quick mode for full quality.


📑 Table of Contents


🚀 Getting Started

✅ Prerequisites

🖥️ Platforms: macOS, Linux, Windows (WSL2)

💳 Claude: Max, Team, or Enterprise recommended

Choose your installation method:

Option A: 🐳 Dev Container (Recommended)

Pre-configured, isolated environment with all tools ready. No system conflicts, consistent across machines, easy cleanup.

Option B: 🍺 Local Installation

Install directly on your system. Requires Homebrew (macOS/Linux/WSL2).

🔧 Installation

Run the following command in your project folder root:

curl -fsSL https://raw.githubusercontent.com/maxritter/claude-codepro/v5.3.1/install.sh | bash

You'll be asked to choose between Dev Container or Local Installation.


📦 What's Inside

♾️ Endless Mode

  • Seamless Continuity - Work on complex features across multiple sessions without losing progress
  • Automatic Handoffs - Context Monitor detects limits and continues seamlessly in new sessions
  • Persistent Memory - Relevant observations automatically carry across all sessions
  • Works Everywhere - Both /spec workflow and Quick Mode benefit from session continuity

📋 Spec-Driven Development

  • Planning - Creates a detailed implementation plan for your review as markdown in docs/plans/
  • Approval - You review, edit if needed, and approve the plan before implementation
  • Implementation - Executes the plan with TDD enforcement and context management
  • Verification - Runs tests, quality checks, and validates completion based on the plan

📚 Modular Rules, Commands & Skills

  • Rules - Best practices for TDD, debugging, context management, and more
  • Commands - /spec for Spec-Driven Development and /sync updating rules
  • Skills - Coding standards for Python, TypeScript, Go, testing, and components
  • Customizable - Add your own rules, commands, and skills that survive updates

🧠 Enhanced Context Capabilities

  • Persistent Memory - Cross-session memory system that automatically ingests context
  • Semantic Search - Local vector store based semantic code search for token-efficient retrieval
  • External Context - Library docs via Context7, web search and scraping via MCP tools
  • Browser Automation - Headless browser automation and testing even within the Dev Container

✅ Quality Automation

  • TDD Enforcer - Pre-edit hook that warns when modifying code without failing tests first
  • Quality Hooks - Language-specific hooks for Python, TypeScript and Go that auto-fix issues
  • LSP Integration - Language servers installed for real-time diagnostics and go-to-definitions
  • Status Line - Live display of context usage, memory status, usage limits, and license info

🛠️ One-Command Installer

  • Automated Container Setup - Isolated Linux environment with pre-configured tools and extensions
  • Extended Language Support - Optionally install extended support for Python, TypeScript & Go
  • Automated Updater - Updates CCP to the latest version when launched over the binary
  • Shell Integration - Auto-configures bash, fish and zsh with ccp alias

🔄 Why Claude CodePro?

Same task. Different results.

| Ad-hoc Prompting | With Claude CodePro | | -------------------------------- | ----------------------------------------------------- | | ❌ Writes code without tests | ✅ TDD enforced — tests first, then implementation | | ❌ No quality checks | ✅ Auto-formatted, linted, type-checked on every edit | | ❌ Inconsistent workflow | ✅ Structured Plan → Implement → Verify cycle | | ❌ Context lost between sessions | ✅ Persistent memory carries observations forward | | ❌ No codebase awareness | ✅ Semantic search understands your code | | ❌ Hope it works | ✅ Verified before marked complete |

The result: Production-grade code you can ship with confidence. Systematic. Tested. Verified.


⚡ Usage

🔄 Sync Rules & Skills

Run /sync to sync custom rules and skills with your codebase. Run it once initially, then anytime after major changes:

ccp > /sync

📋 Spec-Driven Mode

Best for complex features, refactoring, or when you want to review a plan before implementation:

ccp > /spec "Add user authentication with OAuth and JWT tokens"

How it works:

📋 Plan  →  ✅ Approve  →  🔨 Implement  →  🔍 Verify
                                ↑              ↓
                                └── 🔄 Loop ───┘
  1. 📋 Plan — Explores codebase, asks clarifying questions, writes spec to docs/plans/
  2. ✅ Approve — You review and edit the plan, then approve
  3. 🔨 Implement — Executes each task with TDD and quality hooks
  4. 🔍 Verify — Runs tests and checks; loops back if issues found

💬 Quick Mode

Just chat. No plan file, no approval gate. All quality hooks and TDD enforcement still apply.

Best for bug fixes, small improvements, and exploratory work:

ccp > Fix the null pointer bug in user.py

📐 Rules, Commands & Skills

Claude CodePro extends Claude Code with a modular system of rules, commands, and skills:

Managed by CCP (updated on install):

  • .claude/commands/ - spec and sync commands are managed by CCP
  • .claude/rules/standard/ - Best practices and workflows
  • .claude/plugin/skills/ - Coding standards for languages and frameworks

Yours to customize (never touched by updates):

  • .claude/commands/ - Any command besides the standard ones
  • .claude/rules/custom/ - Project-specific rules
  • .claude/skills/ - Any skill besides the standards-* from CCP plugin

When to use what:

| Type | Loaded | Best for | | ------------ | --------------------------------- | --------------------------------------- | | Rules | Every session (always in context) | Guidelines Claude should always follow | | Commands | On demand via /command | Specific workflows or multi-step tasks | | Skills | Dynamically when relevant | Specialized knowledge, coding standards |


🤝 Contributing

Pull Requests - New features, improvements, and bug fixes are welcome. Fun fact: CCP is built with CCP — a self-improving loop where your contributions make the tool that makes contributions better. 🔄

Issues - Found a bug or have a feature request? Open an issue.


📄 License

Try free for 7 days — No signup, no credit card. Just install and go.

Why subscribe? Claude CodePro is actively developed with daily usage in production environments. Your subscription gives you instant access to the latest best practices, workflows, and learnings — a shortcut to state-of-the-art Claude Code development without spending weeks figuring it out yourself.

After your trial, choose the tier that fits your needs here:

| Tier | Includes | |:-----|:---------| | Standard | All features, continuous updates with latest learnings, GitHub support | | Enterprise | Standard + dedicated email support and priority feature requests |

See the LICENSE file for full terms.

五维分析
清晰度8/10
创新性8/10
实用性9/10
完整性8/10
可维护性7/10
优缺点分析

优点

  • 强制执行 TDD 以确保代码可靠
  • 自动化质量检查和格式化
  • 支持多种编程语言
  • 提供持久内存以保持上下文

缺点

  • 需要初始设置和配置
  • 新用户可能需要学习曲线
  • 依赖于 Docker 等外部工具
  • 仅限于特定平台

相关技能

pytorch

S
toolCode Lib / 代码库
92/ 100

“它是深度学习的瑞士军刀,但祝你好运能从47种安装方法里找到那个不会搞崩你系统的那一个。”

agno

S
toolCode Lib / 代码库
90/ 100

“它承诺成为智能体领域的Kubernetes,但得看开发者有没有耐心学习又一个编排层。”

nuxt-skills

S
toolCo-Pilot / 辅助式
90/ 100

“这本质上是一份组织良好的小抄,能把你的 AI 助手变成一只 Nuxt 框架的复读机。”

免责声明:本内容来源于 GitHub 开源项目,仅供展示和评分分析使用。

版权归原作者所有 maxritter.