# Awesome Design MD JP — DESIGN.md for Japanese UI

> Curated Japanese `DESIGN.md` collection (183 sites) to help AI agents generate correct JP typography and UI specs; verified 671★, pushed 2026-05-14.

## Install

Copy the content below into your project:

## Quick Use

```bash
git clone https://github.com/kzhrknt/awesome-design-md-jp
cd awesome-design-md-jp
# Pick a DESIGN.md and copy into your repo root as DESIGN.md
ls design-md | head
```

## Intro

Curated Japanese `DESIGN.md` collection (183 sites) to help AI agents generate correct JP typography and UI specs; verified 671★, pushed 2026-05-14.

**Best for:** Teams building Japanese UIs who need consistent typography, font stacks, and layout rules for agents

**Works with:** Any AI agent/tool that reads Markdown specs (DESIGN.md) alongside project instructions (AGENTS.md/CLAUDE.md)

**Setup time:** 10-20 minutes

### Key facts (verified)

- GitHub: 671 stars · 54 forks · pushed 2026-05-14.
- License: MIT · owner avatar + repo URL verified via GitHub API.
- README-backed entrypoint: `git clone https://github.com/kzhrknt/awesome-design-md-jp && ls design-md | head`.

## Main

- Pick one target’s `DESIGN.md` and treat it as the UI contract for your agent: keep it in-repo and review changes like code.

- Use it to fix CJK typography gaps: README lists Japanese-specific rules (font fallback chains, spacing, kinsoku, OpenType features).

- Pair it with your project instruction file (e.g., `AGENTS.md`) so agents have both behavior rules and visual spec.

- Start with one service’s tokens, then generalize: extract the reusable typography rules into your own DESIGN.md baseline.

### Source-backed notes

- README describes it as a curated `DESIGN.md` collection and links a gallery covering 183 sites.
- README explains why Japanese typography requires different rules (fallback fonts, line height, kinsoku, OpenType).
- README positions `DESIGN.md` alongside agent instruction files like `AGENTS.md` for consistent UI generation.

### FAQ

- **Is this code or a spec library?**: It’s a spec library: README focuses on reusable DESIGN.md documents and typography tokens.
- **How do I use it in a project?**: Copy a chosen DESIGN.md into your repo root and reference it in your agent instructions.
- **What’s a measurable benefit?**: README’s gallery cites 183 site specs; using a concrete spec reduces agent UI guesswork.

## Source & Thanks

> Source: https://github.com/kzhrknt/awesome-design-md-jp
> License: MIT
> GitHub stars: 671 · forks: 54

---

<!-- ZH -->

## Quick Use

```bash
git clone https://github.com/kzhrknt/awesome-design-md-jp
cd awesome-design-md-jp
# Pick a DESIGN.md and copy into your repo root as DESIGN.md
ls design-md | head
```

## Intro

面向日文 UI 的 `DESIGN.md` 策展集合（183 个站点），并提供预览 gallery，帮助 AI agent 按正确排版/字体规则生成 UI；已验证 671★，更新于 2026-05-14。

**Best for:** 做日文 UI，希望给 agent 明确排版/字体/布局规范以生成一致界面的团队

**Works with:** 任何会读取 Markdown 规格说明的 AI agent/tool（与 AGENTS.md/CLAUDE.md 等项目指令并行）

**Setup time:** 10-20 minutes

### Key facts (verified)

- GitHub：671 stars · 54 forks；最近更新 2026-05-14。
- 许可证：MIT；作者头像与仓库链接均已通过 GitHub API 复核。
- README 中可对照的入口命令：`git clone https://github.com/kzhrknt/awesome-design-md-jp && ls design-md | head`。

## Main

- 选择一个站点的 `DESIGN.md` 作为 agent 的 UI 合同：放进仓库并像代码一样审阅变更。

- 用它补齐 CJK 排版缺口：README 列出日文排版要点（字体 fallback、字/行间距、禁则、OpenType 功能等）。

- 与 `AGENTS.md` 等项目指令文件配套，让 agent 同时拿到行为规范与视觉规范。

- 先从一个站点开始，再抽象：把可复用的排版规则沉淀成你自己的 DESIGN.md 基线。

### Source-backed notes

- README 将其描述为 `DESIGN.md` 策展集合，并提供覆盖 183 个站点的 gallery 入口。
- README 解释日文排版为何需要不同规则（字体 fallback、行高、禁则、OpenType 等）。
- README 强调 `DESIGN.md` 可与 `AGENTS.md` 等指令文件并行，帮助 agent 生成一致 UI。

### FAQ

- **这是代码库还是规范库？**：是规范库：README 以可复用的 DESIGN.md 文档与排版 tokens 为核心。
- **怎么在项目里用？**：把选定的 DESIGN.md 复制到项目根目录，并在 agent 指令中要求遵循该规范。
- **有什么可量化的点？**：README 提到 gallery 覆盖 183 个站点规范；有明确规范可减少 agent“瞎猜 UI”。

## Source & Thanks

> Source: https://github.com/kzhrknt/awesome-design-md-jp
> License: MIT
> GitHub stars: 671 · forks: 54


---
Source: https://tokrepo.com/en/workflows/awesome-design-md-jp-design-md-for-japanese-ui
Author: Prompt Lab