Talk2Eve - Main Project

Talk2Eve is a Multi-Framework based chatbot that revolves around using the MITRE ATT&CK knowledge base.

Below, the 5 parts the project needs to function :

• Flutter App : https://github.com/B3LIOTT/Talk2Eve-App
• MiddleWare (Flask) : https://github.com/aymane59/Talk2Eve-MiddlewareAndQueuesHandler
• RabbiMQ (we used a docker)
• Ai (Transformers and MITRE) : https://github.com/Malchemis/Talk2Eve-AI
• MongoDB database (also a docker)

Dockers required

1. RabbitMQ docker:

docker pull rabbitmq

2. MongoDB docker:

docker pull mongo

Configuration

To make this service available from anywhere, we set up a formarding port from a router with a domain name (port 44444) to a Nginx web server (port 443 TLS). Nginx acts as a reverse proxy, transferring requests to a UNIX server running locally on 127.0.0.1:8000: Gunicorn.

1. Create an Nginx server with 127.0.0.1:8000 as reverse proxy and run it with:

sudo systemctl start nginx

2. Run the Gunicorn WSGI with the Flask socketio app:

gunicorn --workers 1 --worker-class eventlet -b 127.0.0.1:8000 wsgi:app

3. Start the AI program:

python main.py

(or set up a service and run it)

4. Run both RabbiMQ and MongoDB dockers with:

docker run <rabbitmq container name>
docker run <mongodb container name>

(or again, as a service)

The architecture