<a href="./i18n/zh/documents/00-基础指南/编程之道.md"><img src="https://img.shields.io/badge/🧭_The%20Art%20of%20Programming-Dao%20Fa%20Shu-orange?style=for-the-badge" alt="The Art of Programming"></a>
AI interpretation link for this repository: [zread.ai/tukuaiai/vibe-coding-cn](https://zread.ai/tukuaiai/vibe-coding-cn/1-overview)
</div>
## 🎲 Foreword
**This is a constantly growing and self-negating project. All current experience and capabilities may become meaningless due to the evolution of AI capabilities. Therefore, please always maintain an AI-first mindset, do not be complacent, all experience may become invalid, view it dialectically🙏🙏🙏**
2. [02-Development Environment Setup](./i18n/zh/documents/01-入门指南/02-开发环境搭建.md) - Copy the prompt to AI and let AI guide you through setting up the environment
3. [03-IDE Configuration](./i18n/zh/documents/01-入门指南/03-IDE配置.md) - Configure VS Code editor
**Vibe Coding** is the ultimate AI pair programming workflow designed to help developers smoothly turn ideas into reality. This guide details the entire process from project conception, tech stack selection, implementation planning to specific development, debugging, and expansion, emphasizing **plan-driven** and **modularization** as the core to prevent AI from losing control and causing project chaos.
> **Core Concept**: *Planning is everything.* Be cautious about letting AI plan autonomously, otherwise your codebase will turn into an unmanageable mess.
**Note**: The experiences shared below are not universally applicable. Please integrate them dialectically with specific scenarios in your practice.
> Further reading: [A Formalization of Recursive Self-Optimizing Generative Systems](./i18n/zh/documents/00-基础指南/A%20Formalization%20of%20Recursive%20Self-Optimizing%20Generative%20Systems.md)
* Feed the newly generated, more powerful artifacts (including new versions of `Ω-Prompt`) back into the system, restarting the continuous evolution to optimize `α-Prompt`.
Through this continuous **recursive optimization loop**, the system achieves **self-transcendence** in each iteration, infinitely approaching the preset **desired state**.
<summary><strong>🧭 Methodology Essentials (Dao · Fa · Shu)</strong></summary>
## 🧭 Dao (The Way)
* **If AI can do it, don't do it manually**
* **Ask AI about everything**
* **Purpose-driven: All actions in the development process revolve around the "purpose"**
* **Context is the first principle of vibe coding; garbage in, garbage out**
* **Systematic thinking, entities, links, functions/purposes, three dimensions**
* **Data and functions are everything in programming**
* **Input, process, output describe the entire process**
* **Ask AI "what?", "why?", "how to do it?" (Golden Circle Rule)**
* **Structure first, then code. Always plan the framework well, otherwise you'll be endlessly paying technical debt**
* **Occam's Razor, don't add code if unnecessary**
* **Pareto Principle, focus on the important 20%**
* **Reverse thinking, first clarify your requirements, then build code backward from requirements**
* **Repeat, try multiple times. If it still doesn't work, open a new window.**
* **Focus, extreme focus can penetrate code. Do one thing at a time (except for divine beings)**
## 🧩 Fa (The Method)
* **One-sentence goal + non-goals**
* **Orthogonality (context-dependent)**
* **Copy, don't write. Don't reinvent the wheel. First ask AI if there's a suitable repository, download and modify it (a new paradigm for glue coding)**
* **Always read the official documentation. First scrape the official documentation and feed it to AI (let AI find tools to download locally)**
* **Decompose modules by responsibility**
* **Interface first, implementation later**
* **Modify only one module at a time**
* **Documentation is context, not an afterthought**
## 🛠️ Shu (The Technique)
* Clearly state: **what can be modified, what cannot be modified**
* For Debugging, only provide: **expected vs actual + minimal reproduction**
* Testing can be handed over to AI, **assertions are human-reviewed**
* When there's too much code, **switch sessions**
* **Consolidate AI-made errors into persistent experience using prompts. If a problem remains unsolved, let AI retrieve this collection of problems and find a solution.**
### Integrated Development Environment (IDE) & Terminal
* [**Visual Studio Code**](https://code.visualstudio.com/): A powerful integrated development environment, suitable for code reading and manual modifications. Its `Local History` plugin is particularly convenient for project version management.
* **Virtual Environment (.venv)**: Highly recommended, allows one-click configuration and isolation of project environments, especially suitable for Python development.
* [**Cursor**](https://cursor.com/): Has captured users' minds and is widely known.
* [**Warp**](https://www.warp.dev/): A modern terminal integrated with AI features, which can effectively improve the efficiency of command-line operations and error troubleshooting.
* [**Neovim (nvim)**](https://github.com/neovim/neovim): A high-performance modern Vim editor with a rich plugin ecosystem, the first choice for keyboard-driven developers.
* [**LazyVim**](https://github.com/LazyVim/LazyVim): A Neovim configuration framework that comes with a full suite of features such as LSP, code completion, and debugging, achieving a balance between out-of-the-box usability and deep customization.
### AI Models & Services
* [**Claude Opus 4.5**](https://claude.ai/new): A powerful AI model, providing services through platforms like Claude Code, and supporting CLI and IDE plugins.
* [**gpt-5.1-codex.1-codex (xhigh)**](https://chatgpt.com/codex/): An AI model suitable for handling large projects and complex logic, available through platforms like Codex CLI.
* [**Droid**](https://factory.ai/news/terminal-bench): Provides CLI access to various models including Claude Opus 4.5.
* [**Kiro**](https://kiro.dev/): Currently provides free Claude Opus 4.5 model access and offers client and CLI tools.
* [**Gemini CLI**](https://geminicli.com/): Provides free access to the Gemini model, suitable for executing scripts, organizing documents, and exploring ideas.
* [**antigravity**](https://antigravity.google/): Currently a free AI service provided by Google, supporting Claude Opus 4.5 and Gemini 3.0 Pro.
* [**AI Studio**](https://aistudio.google.com/prompts/new_chat): A free service provided by Google, supporting Gemini 3.0 Pro and Nano Banana.
* [**Gemini Enterprise**](https://cloud.google.com/gemini-enterprise): Google AI service for enterprise users, currently free to use.
* [**GitHub Copilot**](https://github.com/copilot): An AI code completion tool developed jointly by GitHub and OpenAI.
* [**Kimi K2**](https://www.kimi.com/): A domestic AI model suitable for various general tasks.
* [**GLM**](https://bigmodel.cn/): A domestic large language model developed by Zhipu AI.
* [**Qwen**](https://qwenlm.github.io/qwen-code-docs/zh/cli/): An AI model developed by Alibaba, its CLI tool offers free usage quota.
### Development & Auxiliary Tools
* [**Augment**](https://app.augmentcode.com/): Provides powerful context engine and prompt optimization features.
* [**Windsurf**](https://windsurf.com/): An AI development tool offering free credits for new users.
* [**Ollama**](https://ollama.com/): A local large model management tool, allowing convenient pulling and running of open-source models via the command line.
* [**Mermaid Chart**](https://www.mermaidchart.com/): Used to convert text descriptions into architectural diagrams, sequence diagrams, and other visual charts.
* [**NotebookLM**](https://notebooklm.google.com/): A tool for AI to interpret materials, audio, and generate mind maps.
* [**Zread**](https://zread.ai/): An AI-driven GitHub repository reading tool that helps quickly understand project code.
* [**tmux**](https://github.com/tmux/tmux): A powerful terminal multiplexer, supporting session persistence, split panes, and background tasks, ideal for server and multi-project development.
* [**DBeaver**](https://dbeaver.io/): A universal database management client, supporting various databases with comprehensive features.
### Resources & Templates
* [**Prompt Library (Online Table)**](https://docs.google.com/spreadsheets/d/1ngoQOhJqdguwNAilCl1joNwTje7FWWN9WiI2bo5VhpU/edit?gid=2093180351#gid=2093180351&range=A1): An online table containing a large number of ready-to-use prompts of various types.
* [**Third-party System Prompt Learning Library**](https://github.com/x1xhlol/system-prompts-and-models-of-ai-tools): For learning and referencing system prompts of other AI tools.
* [**Skills Creator**](https://github.com/yusufkaraaslan/Skill_Seekers): A tool that can generate customized Skills according to demand.
* [**Meta Prompts**](https://docs.google.com/spreadsheets/d/1ngoQOhJqdguwNAilCl1joNwTje7FWWN9WiI2bo5VhpU/edit?gid=1770874220#gid=1770874220): Advanced prompts for generating prompts.
* [**General Project Architecture Template**](./i18n/zh/documents/00-基础指南/通用项目架构模板.md): Can be used to quickly set up standardized project directory structures.
* [**Meta-Skill: Skills of Skills**](./i18n/zh/skills/00-元技能/claude-skills/SKILL.md): Meta-skill for generating Skills.
* [**Mobile Remote Vibe Coding**](./i18n/zh/documents/02-方法论/关于手机ssh任意位置链接本地计算机，基于frp实现的方法.md): Achieve mobile SSH remote control of local computers for Vibe Coding based on frp.
### External Tutorials & Resources
* [**Brother Er's Java Advanced Path**](https://javabetter.cn/): Contains detailed configuration tutorials for various development tools.
* [**Virtual Card**](https://www.bybit.com/cards/?ref=YDGAVPN&source=applet_invite): Can be used to register cloud services and other international payment scenarios.
### Community & Communication
* [**Telegram Chat Group**](https://t.me/glue_coding): Vibe Coding Chinese communication group
* [**Telegram Channel**](https://t.me/tradecat_ai_channel): Project updates and information
### Internal Project Documentation
* [**Glue Coding**](./i18n/zh/documents/00-基础指南/胶水编程.md): The holy grail and silver bullet of software engineering, the ultimate evolution of Vibe Coding.
* [**prompts-library Tool Description**](./libs/external/prompts-library/README_CN.md): Supports Excel and Markdown format conversion, including hundreds of curated prompts.
* [**Programming Prompts Collection**](./i18n/zh/prompts/02-编程提示词/): Dedicated prompts for the Vibe Coding process.
* [**System Prompt Construction Principles**](./i18n/zh/documents/00-基础指南/系统提示词构建原则.md): Comprehensive guide to building efficient AI system prompts.
* [**System Prompts Collection**](./i18n/zh/prompts/01-系统提示词/): System prompts for AI development, including multi-version development specifications.
* [**External Resources Aggregation**](./i18n/zh/documents/04-资源/外部资源聚合.md): Curated GitHub repositories, AI tool platforms, prompt resources, and top bloggers.
---
</details>
<details open>
<summary><strong>Encoding Model Performance Grading Reference</strong></summary>
## Encoding Model Performance Grading Reference
It is recommended to only choose top-tier models for complex tasks to ensure optimal results and efficiency.
The core structure of this project `vibe-coding-cn` mainly revolves around knowledge management, AI prompt organization, and automation. Below is a simplified directory tree and descriptions of each part:
```
.
├── CODE_OF_CONDUCT.md # Community code of conduct, regulating contributor behavior.
├── CONTRIBUTING.md # Contribution guide, explaining how to contribute to this project.
├── GEMINI.md # AI assistant context document, including project overview, tech stack, and file structure.
├── LICENSE # Open source license file.
├── Makefile # Project automation scripts for code checking, building, etc.
├── README.md # Main project document, including project overview, usage guide, resource links, etc.
├── .gitignore # Git ignore file.
├── AGENTS.md # AI agent-related documents or configurations.
├── CLAUDE.md # Core behavior rules or configurations for the AI assistant.
│
├── i18n/zh/documents/ # Stores various explanatory documents, experience summaries, and detailed configuration instructions.
│ ├── 00-基础指南/ # Methodologies and principles
│ ├── 04-资源/ # Templates and resources
│ └── 02-方法论/ # Tutorials and guides
│
├── libs/ # Generic library code for project internal modularization.
│ ├── common/ # Common functional modules.
│ │ ├── models/ # Model definitions.
│ │ │ └── __init__.py
│ │ └── utils/ # Utility functions.
│ │ └── backups/ # Internal backup tools.
│ ├── database/ # Database related modules.
│ │ └── .gitkeep # Placeholder file to ensure the directory is tracked by Git.
In a nutshell: Vibe Coding = **Plan-driven + Context fixation + AI pair execution**, turning "from idea to maintainable code" into an auditable pipeline, rather than an unmanageable monolith file.
**What you get**
- A systematic prompt toolchain: `i18n/zh/prompts/01-系统提示词/` constrains AI behavior boundaries, `i18n/zh/prompts/02-编程提示词/` provides full-link scripts for requirement clarification, planning, and execution.
This repository is positioned as a "workflow and prompt" repository rather than a performance-oriented codebase. It is recommended to track the following observable metrics (currently mainly relying on manual recording, scores/traces can be marked in `progress.md`):
| Metric | Meaning | Current Status / Suggestion |
|:---|:---|:---|
| Prompt Hit Rate | Proportion of one-time generation meeting acceptance criteria | To be recorded; mark 0/1 in progress.md after each task |
| Turnaround Time | Time from requirement to first runnable version | Mark timestamps during screen recording, or use CLI timer |
| Change Traceability | Whether context/progress/backup is updated synchronously | Through manual updates; can add git tags/snapshots in backup scripts |
| Example Coverage | Whether there are minimal runnable examples/tests | It is recommended that each example project retains README + test cases |
</details>
---
## 🗺️ Roadmap
```mermaid
gantt
title Project Development Roadmap
dateFormat YYYY-MM
section In Progress (2025 Q4)
Complete Demo GIFs and Example Projects: active, 2025-12, 30d
Add Snapshot and Validation to Backup Script: 2026-02, 10d
section Mid-term (2026 Q2)
Templated Example Project Set: 2026-03, 30d
Multi-model Comparison and Evaluation Baseline: 2026-04, 30d
```
---
## 🎯 Original Repo Translation
> The following content is translated from the original repository [EnzeD/vibe-coding](https://github.com/EnzeD/vibe-coding)
To start Vibe Coding, you only need one of the following two tools:
- **Claude Opus 4.5**, used in Claude Code
- **gpt-5.1-codex.1-codex (xhigh)**, used in Codex CLI
This guide applies to both the CLI terminal version and the VSCode extension version (both Codex and Claude Code have extensions, and the interface is updated).
*(Note: Earlier versions of this guide used **Grok 3**, then switched to **Gemini 2.5 Pro**, and now we are using **Claude 4.5** (or **gpt-5.1-codex.1-codex (xhigh)**))*
*(Note 2: If you want to use Cursor, please refer to [version 1.1](https://github.com/EnzeD/vibe-coding/tree/1.1.1) of this guide, but we believe it is currently not as powerful as Codex CLI or Claude Code)*
- Hand your game idea to **gpt-5.1-codex** or **Claude Opus 4.5**, and let it generate a concise **Game Design Document** in Markdown format, named `game-design-document.md`.
- Review and refine it yourself to ensure it aligns with your vision. It can be rough at first; the goal is to provide AI with the context of the game's structure and intent. Do not over-design; it will be iterated later.
- Let **gpt-5.1-codex** or **Claude Opus 4.5** recommend the most suitable tech stack for your game (e.g., ThreeJS + WebSocket for a multiplayer 3D game), save it as `tech-stack.md`.
- Ask it to propose the **simplest yet most robust** tech stack.
- Open **Claude Code** or **Codex CLI** in the terminal, use the `/init` command, which will read the two .md files you created and generate a set of rules to properly guide the large model.
- **Key: Always review the generated rules.** Ensure the rules emphasize **modularization** (multiple files) and prohibit **monoliths**. You may need to manually modify or supplement the rules.
- **Extremely Important:** Certain rules must be set to **"Always"**, ensuring that AI is forced to read them before generating any code. For example, add the following rules and mark them as "Always":
> ```
> # Important Note:
> # Before writing any code, you must fully read memory-bank/@architecture.md (including full database structure)
> # Before writing any code, you must fully read memory-bank/@game-design-document.md
> # After completing any major feature or milestone, you must update memory-bank/@architecture.md
> ```
- Other (non-Always) rules should guide AI to follow best practices for your tech stack (e.g., networking, state management).
- *If you want the cleanest code and most optimized project, this entire set of rule settings is mandatory.*
- Open **Codex** or **Claude Code** in the VSCode extension, or launch Claude Code / Codex CLI in the project terminal.
- Prompt: Read all documents in `/memory-bank`. Is `implementation-plan.md` completely clear? What questions do you have for me to clarify, to make it 100% clear to you?
- It will typically ask 9-10 questions. After answering all of them, ask it to modify `implementation-plan.md` based on your answers to further refine the plan.
- Open **Codex** or **Claude Code** (extension or terminal).
- Prompt: Read all documents in `/memory-bank`, then execute step 1 of the implementation plan. I will be responsible for running tests. Do not start step 2 until I verify the tests pass. After verification, open `progress.md` to record what you did for future developers, and then add new architectural insights to `architecture.md` to explain the purpose of each file.
- **Always** use "Ask" mode or "Plan Mode" (press `shift+tab` in Claude Code) first, and only let AI execute the step after you are satisfied.
- **Ultimate Vibe:** Install [Superwhisper](https://superwhisper.com), chat freely with Claude or gpt-5.1-codex using voice, no typing needed.
- Commit changes to Git (ask AI if you don't know how).
- Start a new chat (`/new` or `/clear`).
- Prompt: Read all files in memory-bank, read progress.md to understand previous work progress, then proceed to step 2 of the implementation plan. Do not start step 3 until I verify the tests.
- Repeat this process until the entire `implementation-plan.md` is completed.
Congratulations! You have built the basic game! It might still be rough and lack features, but now you can experiment and refine it as much as you want.
- Want fog effects, post-processing, special effects, sound effects? Better planes/cars/castles? Beautiful skies?
- For each major feature you add, create a new `feature-implementation.md`, writing short steps + tests.
- Revert to the previous git commit (`git reset`), try again with new prompts.
- Extremely stuck:
- Use [RepoPrompt](https://repoprompt.com/) or [uithub](https://uithub.com/) to consolidate the entire codebase into one file, then throw it to **gpt-5.1-codex or Claude** for help.
- **Terminal Claude Code / Codex CLI:** Run in VSCode terminal, directly view diffs, feed context, without leaving the workspace.
- **Claude Code's `/rewind`:** One-click rollback to a previous state when iteration goes astray.
- **Custom Commands:** Create shortcut commands like `/explain $parameter` to trigger prompts: "Analyze the code in depth to fully understand how $parameter works. After understanding, tell me, and I will give you the next task." This makes the model load full context before modifying code.
- **Clean Context:** Frequently use `/clear` or `/compact` (to preserve historical conversations).
- **Time-saving trick (use at your own risk):** Use `claude --dangerously-skip-permissions` or `codex --yolo` to completely disable confirmation pop-ups.
- **Small modifications:** Use gpt-5.1-codex (medium)
- **Write top-tier marketing copy:** Use Opus 4.1
- **Generate excellent 2D sprites:** Use ChatGPT + Nano Banana
- **Generate music:** Use Suno
- **Generate sound effects:** Use ElevenLabs
- **Generate videos:** Use Sora 2
- **Improve prompt effectiveness:**
- Add a sentence: "Think slowly, no rush. It's important to strictly follow my instructions and execute perfectly. If my expression is not precise enough, please ask."
- In Claude Code, the intensity of keywords that trigger deep thinking: `think` < `think hard` < `think harder` < `ultrathink`.
- **Q: I'm building an application, not a game. Is the process the same?**
- **A:** Essentially identical! Just replace GDD with PRD (Product Requirements Document). You can also use v0, Lovable, Bolt.new for quick prototyping, then move the code to GitHub, and clone it locally to continue development with this guide.
- **Q: Your aerial combat game's plane model is amazing, but I can't make it with one prompt!**
- **A:** That wasn't one prompt, it was ~30 prompts + a dedicated `plane-implementation.md` file guiding it. Use precise instructions like "cut out space for ailerons on the wings" instead of vague instructions like "make a plane".
- **Q: Why are Claude Code or Codex CLI stronger than Cursor now?**
- **A:** It's entirely a matter of personal preference. We emphasize that Claude Code can better leverage the power of Claude Opus 4.5, and Codex CLI can better leverage the power of gpt-5.1-codex. Cursor's utilization of both is not as good as the native terminal versions. Terminal versions can also work in any IDE, use SSH remote servers, etc., and custom commands, sub-agents, hooks, and other features can significantly improve development quality and speed in the long run. Finally, even if you only have a low-tier Claude or ChatGPT subscription, it's completely sufficient.
- **Q: What if I don't know how to set up a multiplayer game server?**
We warmly welcome contributions in all forms. If you have any ideas or suggestions for this project, please feel free to open an [Issue](https://github.com/tukuaiai/vibe-coding-cn/issues) or submit a [Pull Request](https://github.com/tukuaiai/vibe-coding-cn/pulls).
Before you begin, please take the time to read our [**Contribution Guidelines (CONTRIBUTING.md)**](CONTRIBUTING.md) and [**Code of Conduct (CODE_OF_CONDUCT.md)**](CODE_OF_CONDUCT.md).
---
## 📜 License
This project is licensed under the [MIT](LICENSE) license.
---
<div align="center">
**If this project is helpful to you, please consider giving it a Star ⭐!**
