🌌 Yoda Speak — Galactic Translation Terminal

"Type your words. Translated, they shall be."

A Star Wars-themed translator that converts any text into Yoda's iconic speech pattern. Powered by Llama 4 Scout via Cloudflare Workers AI — no external API keys needed.

🔗 Live demo: yoda-translate.pages.dev

---

✨ Features

• 🤖 AI-powered Yoda speech translation using Llama 4 Scout
• ⚡ Serverless backend via Cloudflare Workers
• 🛡️ Per-IP rate limiting with Cloudflare KV
• 📜 Recent translations history (session-based)
• 🌠 Animated starfield background
• 📋 One-click copy to clipboard
• 📱 Fully responsive design

---

🏗️ Architecture

Browser (Cloudflare Pages)
        │
        │  POST /  { text }
        ▼
Cloudflare Worker  ←── KV (rate limiting)
        │
        │  env.AI.run(llama-4-scout-17b-16e-instruct)
        ▼
Cloudflare Workers AI

---

🚀 Self-hosting

1. Deploy the Worker

1. Go to workers.cloudflare.com and create a new Worker
2. Paste the contents of worker.js
3. Add bindings under Settings → Bindings:
   • Workers AI → variable name: AI
   • KV Namespace → variable name: RATE_LIMIT (create namespace first under Storage & Databases → Workers KV)
4. Deploy

2. Deploy the Frontend

1. Update WORKER_URL in index.html to your Worker URL
2. Go to Workers & Pages → Create → Pages
3. Upload index.html directly
4. Deploy

---

📁 Files

File	Description
index.html	Frontend — single-file UI with starfield, input, and output
worker.js	Cloudflare Worker — rate limiting + Workers AI call
wrangler.toml	Wrangler config — for Git-connected auto-deploys

---

⚙️ Rate Limiting

20 requests per IP per minute, enforced via Cloudflare KV. Resets automatically every 60 seconds.

---

🛠️ Tech Stack

• Frontend: Vanilla HTML/CSS/JS, hosted on Cloudflare Pages
• Backend: Cloudflare Workers (serverless)
• AI: Llama 4 Scout (17B) via Cloudflare Workers AI
• Storage: Cloudflare KV (rate limiting)

---

📄 License

MIT

---

May the Force be with you. ✨