Unisys Innovation Program (UIP) Year 17 - Phase 1
Institution: R V College of Engineering (RVCE), Bangalore
Quantum Guard is a next-generation security gateway designed to intercept and analyze NLP prompts before they reach Large Language Models (LLMs). It integrates Post-Quantum Cryptography (ML-KEM-768) to ensure future-proof data encryption and a BART-based Small Language Model (SLM) for real-time malicious intent scoring and Explainable AI (XAI) feedback.
- Python 3.9+
- Git (properly configured with
.gitignore) - Groq API Key (for LLM processing)
git clone <your-repository-link>
cd Backend_Cl
2. Environment Configuration
Terminal 1:
Create and activate your virtual environment to isolate dependencies:
DOS
python -m venv venv
venv\Scripts\activate
3. Install Dependencies
DOS
pip install -r requirements.txt
4. API Credentials
Create a .env file in the root directory (this file is ignored by Git for security):
Plaintext
GROQ_API_KEY=your_api_key_here
🚦 How to Run (2-Terminal Workflow)
Terminal 1: The Gateway Server (FastAPI)
Run the server with auto-reload enabled. This is ideal for the development phase as it refreshes the server every time you modify main.py, nlp_filter.py, or pqc_layer.py.
DOS
venv\Scripts\activate
uvicorn main:app --reload --port 8000
Wait for the message: Uvicorn running on http://127.0.0.1:8000
Terminal 2: Testing & Client Requests
Keep Terminal 1 running and open a new Command Prompt to simulate a client request:
DOS
curl -X POST "[http://127.0.0.1:8000/process](http://127.0.0.1:8000/process)" ^
-H "Content-Type: application/json" ^
-d "{\"prompt\": \"Analyze the following encrypted PQC packet\"}"
📂 Project Structure
main.py – FastAPI entry point and gateway logic.
nlp_filter.py – SLM scoring engine and XAI reasoning.
pqc_layer.py – ML-KEM-768 encryption/decryption module.
requirements.txt – Project dependencies (FastAPI, Uvicorn, Torch, etc.).
.gitignore – Ensures venv/ and .env are not tracked by Git.