Works with GPT 3.5 turbo and chat history but should also work with Chat GPT 4 you just need the api key for it. Is able to do lists, bullet points, email templates and code blocks. Saved chat history, huge list of prompts added ⌵
Add your Open AI API Key and run it on your local network https://localhost:5000 or with https:// your.domain.com either one works!
- Clone the repo
- Install with Docker CLI / Docker Compose
- Build on Ubuntu or Debian
- Set up a Cloudflare Tunnel for HTTPS optional
- Run on a devcontainer VScode
Project Root
├── TKS-GPT
│ ├── app.py (Flask Backend)
│ ├── .env (Environment Variables)
│ ├── venv (Virtual Environment)
│ └── ...
└── chatbot-ui (React Frontend)
├── package.json (Dependencies and scripts)
├── src
│ ├── App.js (Main component, includes axios API call)
│ └── ...
└── build (Generated by 'npm run build')
Built with python, flask, node.js and react for x86_64 (amd64) architectures and should work on docker desktop also.
git clone https://github.com/bigsk1/TKS-GPT.git
cd TKS-GPT
https://hub.docker.com/r/bigsk1/tks-gpt
Docker
docker run -d -p 5000:5000 --name tks-gpt --env OPENAI_API_KEY=your_openai_api_key --restart always bigsk1/tks-gpt:latest
Docker compose
version: '3.8'
services:
app:
image: bigsk1/tks-gpt:latest
environment:
- OPENAI_API_KEY=<your_openai_api_key>
ports:
- "5000:5000"
restart: unless-stopped
security_opt:
- no-new-privileges:true
You can use docker cli in same folder as the Dockerfile
docker build -t your_image_name .
docker run -d -p 5000:5000 --name tksgpt-container --env OPENAI_API_KEY=ENTER_YOUR_OPEN_AI_APIKEY_HERE your_image_name
or
In the docker-compose.yml file add your Open AI Api key
version: '3.8'
services:
app:
build: .
image: your_custom_image_name
restart: unless-stopped
environment:
- OPENAI_API_KEY=your_openai_api_key
ports:
- "5000:5000"
and then
docker-compose up -d
Your app will be running on https://localhost:5000 use it as is or add ssl both work with no code change needed.
THERE ARE TWO PARTS TO YOUR PROJECT A BACKEND AND A FRONTEND
Active Virtual Enviroment ( optional but recommended )
For Python, you'll need to create a virtual environment. In your project directory, run:
python3 -m venv venv
Activate the virtual environment:
source venv/bin/activate
Install Python dependencies:
With the virtual environment activated, install the required Python packages using the requirements.txt file:
pip install -r requirements.txt
Exit out of virtual enviroment and back to the TKS-GPT folder type deactivate
Install Node.js and npm:
Make sure Node.js and npm are installed on your system. You can check if they are installed by running:
node -v
npm -v
You need Node version 18+
If not installed, you can download them from the official Node.js website https://nodejs.org/en/download/
See here https://github.com/nodesource/distributions
Install frontend dependencies:
Node (v19.x) to current
sudo apt update
sudo apt install -y curl
curl -fsSL https://deb.nodesource.com/setup_current.x | sudo -E bash - &&\
sudo apt-get install -y nodejs
sudo apt-get install -y build-essential
Node (v19.x) to current
sudo apt install curl
curl -fsSL https://deb.nodesource.com/setup_current.x | bash - &&\
apt-get install -y nodejs
run as root on debian
apt-get install -y build-essential
Navigate to the frontend folder (the folder containing the package.json file, e.g., chatbot-ui), and install the required npm packages:
cd chatbot-ui
npm install
To connect the front-end to the back-end
Install Axios in your chatbot-ui folder:
npm install axios
Set up environment variables:
Make sure you have a .env file in (TKS-GPT folder) with the required environment variables (such as OPENAI_API_KEY).
After completing these steps, you should be able to run both the backend and frontend servers and start using your chatbot application.
TKS-GPT
├── app.py
├── chatbot-ui
│ ├── build
│ │ └── ...
│ ├── node_modules
│ │ └── ...
│ ├── public
│ │ └── ...
│ ├── src
│ │ ├── App.css
│ │ ├── App.js
│ │ ├── App.test.js
│ │ ├── index.css
│ │ ├── index.js
│ │ ├── logo.svg
│ │ ├── MessageContent.js
│ │ ├── Prompts.js
│ │ ├── reportWebVitals.js
│ │ ├── SavedChats.css
│ │ ├── SavedChats.js
│ │ ├── setupTests.js
│ │ └── ...
│ ├
│ ├── .gitignore
│ ├── package.json
│ └── package-lock.json
├── .dockerignore
├── Dockerfile
├── docker-compose.yml
├── .env
├── flask_app.log
├── .gitignore
├── requirements.txt
└── venv
└── ...
Backend: The Flask backend is in the TKS-GPT directory. It serves the chatbot API and the React frontend (static files in chatbot-ui/build).
Frontend: The React frontend is in the chatbot-ui directory. It contains the main App.js file where the axios API call is made to the Flask backend.
Axios: Axios is used in the App.js file to make an HTTP POST request to the Flask backend's /chat endpoint. The request contains the user's message, and the response from the server contains the AI-generated message.
React: React is used for building the frontend user interface. The project was created using Create React App, and it uses components and state management with hooks (e.g., useState).
Node.js: Node.js is used as the underlying runtime environment for running the development server (npm start) and for building the production-ready static files (npm run build). It also manages the dependencies through package.json
The frontend and backend communicate via an API call made with axios from the App.js file. When the user enters a message and clicks the "Send" button, the frontend sends an HTTP POST request to the backend's /chat endpoint. The backend processes the request, interacts with the GPT model, and returns the AI-generated response, which the frontend then displays.
cd TKS-GPT/chatbot-ui
npm run build
In the TKS-GPT project folder
source venv/bin/activate
python3 app.py
On ubuntu you can use systemd to create a service to keep running even after a restart
make file for systemd service add your own user, group and path's
sudo nano /etc/systemd/system/tkschat.service
[Unit]
Description=TKS Chat Bot
After=network.target
[Service]
User=farlab
Group=farlab
WorkingDirectory=/home/farlab/Documents/EthicalAutonomy/TKS-GPT
Environment="/home/farlab/Documents/EthicalAutonomy/TKS-GPT/venv/bin"
ExecStart=/home/farlab/Documents/EthicalAutonomy/TKS-GPT/venv/bin/python app.py
Restart=on-failure
RestartSec=5s
[Install]
WantedBy=multi-user.target
Save and exit the file.
Reload the systemd daemon to register your new service
sudo systemctl daemon-reload
Enable your service to start at boot:
sudo systemctl enable tkschat.service
Start your service:
sudo systemctl start tkschat.service
Now, your Flask app will automatically start when the system boots up. To check the status of your service, use the following command:
sudo systemctl status tkschat.service
You can also stop, restart, or disable the service using systemctl commands:
sudo systemctl stop tkschat.service
sudo systemctl restart tkschat.service
sudo systemctl disable tkschat.service
check systemd logs for errors
journalctl -u tkschat.service
Your App will be on https://YOUR-LOCAL-IP-ADDRESS:5000
The easiest way is to use the UI in zero trust and make a tunnel and assign it a domain or subdomain. Allow your domain as origin, POST and GET methods, all headers.
Want to make updates and changes and using systemd?
first stop systemd server for flask using
sudo systemctl stop tkschat.service
in /TKS-GPT/chatbot-ui
npm run build
once built can enabled again using
sudo systemctl restart tkschat.service
git clone https://github.com/bigsk1/TKS-GPT.git
cd TKS-GPT
Open in Dev Container and when asked in terminal enter Open Ai Api Key, browse to localhost:5000
Read more about how this app was built using Chat-GPT https://bigsk1.github.io/posts/building-an-app-with-chat-gpt/