website-cloner

Website Cloner

You are about to clone a website into a fully self-contained static site — pure HTML, CSS, and JavaScript with zero framework dependencies. The output is a folder of static files that can be opened directly in a browser, served by any static server, or migrated into any framework later.

Interactive Input

If $ARGUMENTS is empty or incomplete, ask the user interactively. Do NOT proceed without the required information. Ask questions one at a time — don't overwhelm with a long form.

Step 1: Target URL

If no URL was provided in $ARGUMENTS, ask:

Q: 要克隆哪个网站？请提供目标 URL。

• Option A: 输入 URL（推荐）— "提供目标网址，例如 https://example.com"
• Option B: 仅本地参考 — "我没有 URL，只有本地已下载的静态文件"

If the user chooses B (no URL), skip to Step 2 but note that extraction will rely entirely on local files — no live fetch possible.

Validate the URL format. If invalid, ask again.

Step 2: Local Reference (Optional)

Ask:

Q: 是否有已下载的静态站点作为参考？

• Option A: 有（推荐）— "提供本地目录路径"
• Option B: 没有 — "直接克隆，无需参考"

If the user provides a path, verify the directory exists. If not, ask again or offer to skip.

Step 3: Output Directory Name

If --dir was not provided in $ARGUMENTS, suggest a default derived from the hostname and ask:

Q: 输出目录叫什么名字？

• 显示建议名称（如 example-com），让用户确认或自定义
• 用户可直接回车确认默认值，或输入新名称

If the directory already exists, ask whether to overwrite or pick a new name.

Step 4: Confirm Summary

Before starting, show a summary and ask for confirmation:

即将克隆：

• 目标 URL: https://example.com
• 本地参考: ~/Downloads/example-site（如有）
• 输出目录: ./example-com/

确认开始？

Only proceed after the user confirms.

---

Once all inputs are gathered, the workflow proceeds with the collected information.

Input Modes

Based on gathered input, this skill operates in three modes:

1. URL only — Fetch the live site, extract everything, rebuild as static files.
2. URL + local reference — The user has already downloaded a static version (e.g., via browser "Save Page" or wget). Use it as a reference to cross-check assets and structure, fill in gaps the live fetch might miss.
3. Local reference only — No live URL; rebuild from the downloaded files alone.