ThrottlePoint 🚦🧠

A smart reverse proxy with built-in rate limiting — soon to be AI-powered.

🛠 What is ThrottlePoint?

ThrottlePoint is a lightweight reverse proxy server written in Go, designed to:

• Intercept and forward HTTP requests
• Apply configurable rate limiting (currently using the Token Bucket algorithm)
• Be easily extendable for real-time AI-based intent analysis and dynamic rate limiting

Think of it as your traffic cop at the API junction — smart, fast, and soon, intuitive.

---

✅ Features (Implemented)

🌐 Reverse Proxy

• Forwards any incoming request (/*) to a backend server.
• Preserves method, headers, and query params.
• Pipes the response back to the client.

⛔ Rate Limiting (Token Bucket)

• Rate limit per IP address.
• Configurable token rate and capacity.
• Efficient and thread-safe (uses sync.Map and per-bucket mutex).
• Background cleanup of inactive buckets.

⚙️ Configurable Backend

• Uses environment variable BACKEND_URL to forward requests (default: http://localhost:3000).
• Easy to adapt for different backends.

---

🧠 Upcoming Features (The Real Deal)

🤖 AI-Enhanced Intent Prediction (Core Vision)

• Analyze request metadata (IP, headers, payload).
• Predict request “intent” or risk level using lightweight ML models.
• Smart categorization (e.g., normal traffic vs. scrapers vs. potential abuse).

📊 Dynamic Rate Limiting

• Adapt limits in real-time based on intent scores.
• Trusted users = more leniency.
• Suspicious traffic = throttled or blocked.

🔄 Feedback Loop

• Log request outcomes.
• Fine-tune AI model/rules continuously for improved predictions.
• Enable reinforcement-style learning.

📈 Dashboard & Metrics (Stretch)

• Web dashboard to view request logs, AI decisions, and live traffic stats.
• Manual overrides and real-time debugging.

---

🚀 Quickstart

# clone the repo
git clone https://github.com/asutosh2203/throttle-point.git
cd throttle-point

# create .env file
echo "BACKEND_URL=<YOUR BACKEND URL>" > .env

# run the app
go run main.go

---

📁 Project Structure

.
├── main.go                          # Entry point
├── handlers/
│ └── proxy.go                      # Request forwarding logic
├── middleware/
│ └── rateLimiter.go              # Rate limiting middleware
│ └── token_bucket.go          # Token Bucket middleware
├── go.mod
├── .env                                 # Backend target URL
├── .gitignore
└── README.md

---

🤝 Contributing

AI brains welcome. If you’ve got ideas on risk scoring, model integration, or smarter rate control — hop in!

---

⚡️ Inspiration

Born from frustration with blunt rate-limiters and a love for AI-enhanced infrastructure.

---

Maintained by Asutosh
Drop a ⭐ if you like where this is going!