MahanaicoachMahanaicoach/ai-site-cloner/.github/copilot-instructions.md

ai-site-cloner

Copilot instructions31 starsChanged 3 months agoMITView on GitHub

Template for cloning any multi-page website into this Next.js codebase, then rebranding it as your own. Two commands drive everything: /clone-website <url> and /restyle. Two workflows, each with a single source of truth in .claude/skills/<name>/SKILL.md: Where your tool finds them (all generated from the same source): The extraction scripts are plain Node CLIs any agent can run; Chromium auto-installs on first use (or npm run setup). No subagent support in your tool? Build sections sequentially from the same lint-passing specs.

We split this copilot-instructions.md into the rules it gives an agent

2 rules. About 0 are common, 0 are distinctive, and 2 are very specific to this project.

≈0
Common · estimate
Many files like this have a rule of this kind.
≈0
Distinctive · estimate
This author's own choices. Worth a look.
2
Very specific
Tied to their project's names, files or tools. Probably skip.

Common
Distinctive
Very specific

Very specific:
All extraction numbers come from scripts/ (computed styles, real column counts, pixel diffs) — never estimated by eye.
line 27
Very specific:
After editing either, run node scripts/sync-skills.mjs to regenerate every per-tool copy.
line 68

≈ “Common” is an estimate: at least 1 in 5 of the 3,669 Copilot instructions files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.

Files

copilot-instructions.md

<!-- GENERATED from AGENTS.md by scripts/sync-skills.mjs — edit the source,
     then run: node scripts/sync-skills.mjs -->

ai-site-cloner

Template for cloning any multi-page website into this Next.js codebase, then rebranding it as your own. Two commands drive everything: /clone-website <url> and /restyle.

Tech Stack

Next.js 16 — App Router, React 19, TypeScript strict
Tailwind CSS v4 + shadcn/ui (base-ui primitives), oklch tokens in src/app/globals.css
Lucide React icons (replaced by extracted SVGs in src/components/icons.tsx during cloning)
Playwright (devDependency) powers all extraction/QA scripts

Commands

npm run dev / npm run build / npm run lint / npm run typecheck
npm run check — lint + typecheck + build (must pass before any commit)

Architecture Rules

Content lives in src/data/*.ts, never hardcoded in components. Components take props. This is what makes /restyle possible — copy swaps without touching layout.
Pages: src/app/<route>/page.tsx wires data → components. Shared header/footer mount in layouts.
docs/research/ holds extraction artifacts (JSON, specs, manifest) — auditable, never hand-edited except spec files.
docs/research/manifest.json is the pipeline state. Update it at every stage transition; check it first on any clone-related task (node scripts/manifest.mjs status).
Specs (docs/research/components/**/*.spec.md) must pass node scripts/lint-spec.mjs before any builder agent is dispatched.
All extraction numbers come from scripts/ (computed styles, real column counts, pixel diffs) — never estimated by eye.

Code Style

TypeScript strict, no any
Named exports, PascalCase components, camelCase utils
Tailwind utility classes; arbitrary values (text-[17px]) when the extracted value doesn't match Tailwind's scale
2-space indentation

Skills / Commands

Two workflows, each with a single source of truth in .claude/skills/<name>/SKILL.md:

Command	Purpose
`/clone-website <url>`	Crawl, extract, spec, build, and pixel-diff QA a full multi-page clone
`/restyle`	Swap colors, fonts, logo, and copy for your brand — layout stays intact

Where your tool finds them (all generated from the same source):

Claude Code: .claude/skills/<name>/SKILL.md
Codex CLI: .codex/skills/<name>/SKILL.md
GitHub Copilot coding agent: .github/skills/<name>/SKILL.md
Cursor: .cursor/commands/<name>.md
Windsurf: .windsurf/workflows/<name>.md
Gemini CLI: .gemini/commands/<name>.toml
OpenCode: .opencode/commands/<name>.md
Continue: .continue/commands/<name>.md
Augment Code: .augment/commands/<name>.md
Amazon Q Developer CLI: .amazonq/cli-agents/<name>.json
Cline / Roo Code, Aider, anything else: open .claude/skills/<name>/SKILL.md and follow it as a prompt.

The extraction scripts are plain Node CLIs any agent can run; Chromium auto-installs on first use (or npm run setup). No subagent support in your tool? Build sections sequentially from the same lint-passing specs.

Ethics

Clone sites you own or have permission to rebuild (platform migration, lost source, learning). Never for phishing, impersonation, or passing off someone's design as your own.

Keeping Agent Configs in Sync

.claude/skills/*/SKILL.md and this file (AGENTS.md) are the only sources of truth for per-tool configs.
After editing either, run node scripts/sync-skills.mjs to regenerate every per-tool copy.
Generated files are committed; CI runs node scripts/sync-skills.mjs --check and fails if any are stale.

Download now (ZIP)

Where files are headingNew Copilot instructions files have more rules about branches and pull requests, and fewer about code style.
Of the Copilot instructions files we read, the 2,297 first seen in the last six months, against the older ones: the kind of rule that gained most share, and the kind that lost most. About files like this in general, not this one, and an estimate, since a kind is a pattern in the words.
In this repository7 other files it gives its agents.
- AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- .cursor/rules/project.mdcCursor rule
- clone-websiteSkill
- restyleSkill
- clone-websiteSkill
- restyleSkill

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

It worked
Partly
It didn't

What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

Check your setup, free