|
| 1 | +<div align="center"> |
| 2 | + |
| 3 | +# Vibe Coding Guide |
| 4 | + |
| 5 | +**Move AI coding from prompt experiments to a reviewable engineering workflow.** |
| 6 | + |
| 7 | +Language: English | [中文](./README.md) |
| 8 | + |
| 9 | +[Start on the Website](https://lling0000.github.io/Vibe_coding_guide/) · |
| 10 | +[English Guide](./vibe-coding-guide-en.md) · |
| 11 | +[中文教程](./vibe-coding-guide-zh.md) · |
| 12 | +[Roadmap](./docs/roadmap.md) |
| 13 | + |
| 14 | +[English PDF](./vibe-coding-guide-en.pdf) · |
| 15 | +[中文 PDF](./vibe-coding-guide-zh.pdf) · |
| 16 | +[Contributing](./CONTRIBUTING.md) |
| 17 | + |
| 18 | +[](https://lling0000.github.io/Vibe_coding_guide/) |
| 19 | +[](https://github.com/Lling0000/Vibe_coding_guide/actions/workflows/docs.yml) |
| 20 | +[](./vibe-coding-guide-en.md) |
| 21 | +[](./vibe-coding-guide-zh.md) |
| 22 | +[](https://github.com/topics/ai-coding) |
| 23 | +[](./LICENSE-CONTENT) |
| 24 | +[](./LICENSE) |
| 25 | + |
| 26 | +</div> |
| 27 | + |
| 28 | +--- |
| 29 | + |
| 30 | +## What This Is |
| 31 | + |
| 32 | +AI coding breaks down when it stays at the prompt-to-code level: vague requests produce plausible diffs, long chats lose context, parallel agents collide, and "looks right" replaces verification. |
| 33 | + |
| 34 | +Vibe Coding Guide solves that gap by treating AI coding as an engineering workflow: a repeatable loop for specifying work, feeding agents the right context, reviewing changes, testing behavior, committing safely, and handing work off. It is not a collection of magic prompts; it is the operating system around AI-assisted development: |
| 35 | + |
| 36 | +- specs that turn vague intent into a contract |
| 37 | +- durable project context through `AGENTS.md` / `CLAUDE.md` |
| 38 | +- context-window hygiene, compression, handoff, and reset habits |
| 39 | +- subagents, workflow patterns, and multi-agent coordination |
| 40 | +- git worktrees for parallel agent development |
| 41 | +- reusable skills for repeated tasks |
| 42 | +- CI, tests, and review habits for agent-written code |
| 43 | + |
| 44 | +The goal is not to "let AI code for you." The goal is to become a stronger operator of AI coding agents. |
| 45 | + |
| 46 | +## Start Here |
| 47 | + |
| 48 | +| If you are... | Start with | What you should do first | |
| 49 | +|---|---|---| |
| 50 | +| Skimming before starring | [Website](https://lling0000.github.io/Vibe_coding_guide/) | Open the 16-day Feynman checklist and read the first day | |
| 51 | +| Reading in English | [vibe-coding-guide-en.md](./vibe-coding-guide-en.md) | Read chapters 1-5 before copying any workflow | |
| 52 | +| Reading in Chinese | [README.md](./README.md) | Use the Chinese entry page, then jump into the full guide | |
| 53 | +| Using Codex, Claude Code, Cursor, or Aider | Chapters 1-5 | Write one real spec and one project `AGENTS.md` / `CLAUDE.md` | |
| 54 | +| Running multiple agents or sessions | Chapters 6-9 | Learn subagents, workflow patterns, `.gitignore`, and worktrees | |
| 55 | +| Rolling this out to a team | Chapters 10-13 | Turn repeated work into skills and add CI/testing guardrails | |
| 56 | +| Auditing your own habits | Chapters 14-16 | Compare your current workflow against the anti-pattern checklist | |
| 57 | + |
| 58 | +## The Engineering Loop |
| 59 | + |
| 60 | +```mermaid |
| 61 | +flowchart LR |
| 62 | + A["Spec"] --> B["Context"] |
| 63 | + B --> C["Agent Plan"] |
| 64 | + C --> D["Implementation"] |
| 65 | + D --> E["Diff Review"] |
| 66 | + E --> F["Tests and CI"] |
| 67 | + F --> G["Commit"] |
| 68 | + G --> H["Docs and Handoff"] |
| 69 | + H --> A |
| 70 | +``` |
| 71 | + |
| 72 | +Vibe Coding works when this loop is explicit. Every phase should leave evidence in files, diffs, command output, tests, or commits. |
| 73 | + |
| 74 | +## How This Differs From a Prompt Guide |
| 75 | + |
| 76 | +| Ordinary prompt guide | Vibe Coding Guide | |
| 77 | +|---|---| |
| 78 | +| Optimizes a single request | Optimizes the whole engineering loop around the request | |
| 79 | +| Asks "what should I type?" | Asks "what system makes agent work reviewable?" | |
| 80 | +| Focuses on phrasing | Focuses on specs, context, files, git, CI, tests, and handoff | |
| 81 | +| Measures success by a plausible answer | Measures success by a diff that satisfies acceptance checks | |
| 82 | +| Treats chat history as memory | Moves durable knowledge into `AGENTS.md`, specs, docs, and skills | |
| 83 | +| Relies on manual recovery | Uses worktrees, commits, resets, and review gates to recover cleanly | |
| 84 | + |
| 85 | +## Core Concepts |
| 86 | + |
| 87 | +| Concept | Plain meaning | Why it matters | |
| 88 | +|---|---|---| |
| 89 | +| Spec | The contract for what the agent should change | Prevents vague work and gives you acceptance criteria | |
| 90 | +| Context | The files, rules, history, and examples the agent can use | Keeps the agent focused on the right local reality | |
| 91 | +| Agent plan | The proposed route before edits begin | Lets you stop bad direction before it becomes code | |
| 92 | +| Subagent | A separate worker for isolated search, review, or analysis | Keeps the main session from filling with unrelated context | |
| 93 | +| Workflow | A deterministic sequence of steps around the agent | Makes collaboration repeatable instead of improvised | |
| 94 | +| Worktree | A separate checkout for parallel work | Lets multiple agents move without stepping on each other | |
| 95 | +| Skill | A reusable task procedure | Turns repeated prompts into maintained operating knowledge | |
| 96 | +| CI/testing | Automated evidence that work still behaves | Replaces "looks right" with repeatable verification | |
| 97 | + |
| 98 | +## Learning Paths |
| 99 | + |
| 100 | +**30-minute orientation** |
| 101 | + |
| 102 | +1. Read chapters 1-2 to understand the role shift from typing code to directing agent attention. |
| 103 | +2. Skim chapter 3 and write a tiny `AGENTS.md` / `CLAUDE.md` for one repository. |
| 104 | +3. Open the website checklist and write a three-minute explanation of the core loop. |
| 105 | + |
| 106 | +**First real project** |
| 107 | + |
| 108 | +1. Write a lightweight spec for a small change. |
| 109 | +2. Ask the agent for a plan before edits. |
| 110 | +3. Review the diff, run verification, and commit. |
| 111 | +4. Record one mistake or local convention back into project docs. |
| 112 | + |
| 113 | +**Parallel agent practice** |
| 114 | + |
| 115 | +1. Read chapters 6-9. |
| 116 | +2. Give broad discovery to a subagent or separate session. |
| 117 | +3. Put risky implementation in a dedicated git worktree. |
| 118 | +4. Merge only after review and tests pass. |
| 119 | + |
| 120 | +**Team adoption** |
| 121 | + |
| 122 | +1. Read chapters 10-13. |
| 123 | +2. Convert one repeated workflow into a skill. |
| 124 | +3. Add CI rules to the project agent guide. |
| 125 | +4. Build a small case library for evaluating agent behavior. |
| 126 | + |
| 127 | +## Chapter Map |
| 128 | + |
| 129 | +| Phase | Chapters | Outcome | |
| 130 | +|---|---:|---| |
| 131 | +| Foundations | 1-5 | Write better specs, maintain project context, and manage long sessions | |
| 132 | +| Coordination | 6-9 | Use subagents, workflow patterns, `.gitignore`, and worktrees | |
| 133 | +| Reuse and guardrails | 10-13 | Create skills, separate prompt layers, add CI/CD, and test agent behavior | |
| 134 | +| Judgment | 14-16 | Build durable habits and spot failure modes early | |
| 135 | + |
| 136 | +| Chapter | Topic | |
| 137 | +|---:|---| |
| 138 | +| 1 | What Vibe Coding is and how your role changes | |
| 139 | +| 2 | Specs as the starting point for useful agent work | |
| 140 | +| 3 | What belongs in `AGENTS.md` / `CLAUDE.md` | |
| 141 | +| 4 | Cold-starting new and inherited projects | |
| 142 | +| 5 | Context management, compression, handoffs, and resets | |
| 143 | +| 6 | Subagents and context isolation | |
| 144 | +| 7 | Workflow patterns and multi-agent collaboration | |
| 145 | +| 8 | `.gitignore` as repository hygiene | |
| 146 | +| 9 | Git worktrees for parallel agent development | |
| 147 | +| 10 | Skills as reusable task workflows | |
| 148 | +| 11 | System prompts vs user prompts | |
| 149 | +| 12 | CI/CD guardrails for agent-written code | |
| 150 | +| 13 | Testing ordinary code and testing agent behavior | |
| 151 | +| 14 | Advanced operating principles | |
| 152 | +| 15 | A complete multi-day workflow example | |
| 153 | +| 16 | Anti-patterns checklist | |
| 154 | + |
| 155 | +## Repository Structure |
| 156 | + |
| 157 | +```text |
| 158 | +. |
| 159 | +├── index.html # GitHub Pages checklist and reader entry |
| 160 | +├── assets/ # Site styles, script, and visual asset |
| 161 | +├── docs/ |
| 162 | +│ └── roadmap.md # Public roadmap and contribution priorities |
| 163 | +├── .github/workflows/ |
| 164 | +│ └── docs.yml # Docs-only sanity checks for local links |
| 165 | +├── README.md # Chinese default front door |
| 166 | +├── README.en.md # English front door |
| 167 | +├── README.zh-CN.md # Chinese compatibility front door |
| 168 | +├── README_zh.md # Legacy Chinese entry link |
| 169 | +├── CONTRIBUTING.md # Contribution guide |
| 170 | +├── vibe-coding-guide-en.md # Full English tutorial |
| 171 | +├── vibe-coding-guide-en.pdf # English PDF |
| 172 | +├── vibe-coding-guide-zh.md # Full Chinese tutorial |
| 173 | +└── vibe-coding-guide-zh.pdf # Chinese PDF |
| 174 | +``` |
| 175 | + |
| 176 | +## Project Status |
| 177 | + |
| 178 | +This is a documentation-first repository. The full bilingual guides and PDFs are already included, and the GitHub Pages site provides a 16-day Feynman-style learning checklist with local browser progress. |
| 179 | + |
| 180 | +There is currently no package manager setup because the site is plain static HTML/CSS/JavaScript. The CI added here only checks repository-facing documentation links; it does not build or publish a package. |
| 181 | + |
| 182 | +## Contributing |
| 183 | + |
| 184 | +Good contributions make the guide more concrete, verifiable, and useful in real workflows. See [CONTRIBUTING.md](./CONTRIBUTING.md) and [docs/roadmap.md](./docs/roadmap.md) before opening a pull request. |
| 185 | + |
| 186 | +## License |
| 187 | + |
| 188 | +This repository uses a dual-license model: |
| 189 | + |
| 190 | +- Documentation, guides, PDFs, and written instructional content are licensed under [CC BY 4.0](./LICENSE-CONTENT). |
| 191 | +- Website code, scripts, styles, and other software source files are licensed under the [MIT License](./LICENSE). |
| 192 | + |
| 193 | +If a file does not say otherwise, use the license that matches its primary purpose: written guide content uses CC BY 4.0; executable or reusable code uses MIT. |
0 commit comments