
uholli/ai-website-cloner 是从 https://github.com/uholli/ai-website-cloner. 导入的 Agent Skill 候选人。
概览
uholli/ai-website-cloner 是从 https://github.com/uholli/ai-website-cloner. 导入的 Agent Skill 候选人。
概述(AI 翻译)
一个面向生产环境的入门工具,用于用AI编码代理重建活页网站,作为类型化Next.js应用。该模板包含严格的TypeScript工具链、可选的Redis支持会话持久性,以及多平台代理技能,确保/clone-website工作流程在团队间保持可重复性。
从你自己的副本开始。 使用 GitHub ** 使用此模板**创建独立仓库。不要提交生成的克隆输出到上游模板。
目录
- 为什么选择这个模板
- 系统架构
- 克隆工作流程
- 特色
- 要求
- 安装
- 配置
- 发展
- 测试
- 项目结构
- 故障排除
- 贡献
- 常见问题解答
- 许可证
为什么要选这个模板
大多数网站重建在设计提取和组件实现之间的交接过程中都会失败。本仓库标准化了这一边界:
| 关切 | 方法 |
|---|---|
| 代理指令 | AGENTS.md中的单一源,同步至10+平台 |
| 设计忠实度 | 组件规格中计算出的CSS值见docs/research/ |
| 运行时质量 | 严格的TypeScript、ESLint、Vitest和CI验证 |
本页由 AI 翻译,技术细节请以原文为准。
README
以下为原文 README 节选,完整内容请查看原文。
Website Cloner — Next.js Agent Template
A production-oriented starter for rebuilding live websites as typed Next.js applications with AI coding agents. The template ships a strict TypeScript toolchain, optional Redis-backed session persistence, and multi-platform agent skills so /clone-website workflows stay reproducible across teams.
Start from your own copy. Use GitHub Use this template to create a separate repository. Do not commit generated clone output back to this upstream template.
Table of contents
- Why this template
- System architecture
- Clone workflow
- Features
- Requirements
- Installation
- Configuration
- Development
- Testing
- Project structure
- Troubleshooting
- Contributing
- FAQ
- License
Why this template
Most website rebuilds fail in the handoff between design extraction and component implementation. This repository standardizes that boundary:
| Concern | Approach |
|---|---|
| Agent instructions | Single source in AGENTS.md, synced to 10+ platforms |
| Design fidelity | Component specs with computed CSS values in docs/research/ |
| Runtime quality | Strict TypeScript, ESLint, Vitest, and CI validation |
| Session continuity | Optional Redis cache for multi-step clone pipelines |
| Deployment | Standalone Next.js output with Docker Compose |
System architecture
flowchart TB
subgraph agents [AI Coding Agents]
CC[Claude Code]
CX[Codex / Cursor / Copilot]
Others[Gemini / Windsurf / ...]
end
subgraph template [Template Repository]
SKILL["/clone-website skill"]
AGENTS[AGENTS.md rules]
APP[Next.js App Router]
CACHE[Cache layer]
API["/api/cache/*"]
end
subgraph persistence [Optional Persistence]
MEM[(In-memory store)]
REDIS[(Redis)]
end
subgraph output [Generated Clone]
COMP[React components]
ASSETS[public/images + seo]
SPECS[docs/research/components]
end
agents --> SKILL
SKILL --> AGENTS
AGENTS --> APP
APP --> API
API --> CACHE
CACHE --> MEM
CACHE --> REDIS
SKILL --> SPECS
SPECS --> COMP
SKILL --> ASSETS
Cache layer
sequenceDiagram
participant Agent
participant API as /api/cache/sessions
participant Service as CacheService
participant Backend as Memory or Redis
Agent->>API: POST session metadata
API->>Service: saveCloneSession()
Service->>Backend: SET session:{id}
Backend-->>Service: OK
Service-->>API: session record
API-->>Agent: 201 Created
Agent->>API: GET ?id=session-1
API->>Service: getCloneSession()
Service->>Backend: GET session:{id}
Backend-->>Service: JSON payload
Service-->>API: CloneSessionRecord
API-->>Agent: 200 OK
Clone workflow
推荐工具
换一个关键词,或者移除筛选条件。
安装
npx skillfish add uholli/ai-website-cloner