BioBot 🧪🤖

BioBot is a web application and chatbot designed for laboratory automation, particularly with liquid handling robots. It helps users interact with an AI assistant to get guidance, protocols, and Python code for automating lab workflows.

---

Features

• User Registration and Login: Secure authentication system with hashed passwords.
• Chat Interface: Talk to the BioBot assistant with context-aware replies.
• Code Generation: Generates Python code for lab automation protocols.
• RAG (Retrieval-Augmented Generation): Fetches relevant documentation and generates answers for lab-related queries.
• Dockerized Deployment: Easy to deploy with Docker and Docker Swarm, supporting secrets for API keys.
• Persistent Chat History: Stores user chats and allows multiple concurrent conversations.
• Secure API Key Handling: Supports API key retrieval via Docker secrets.

---

Tech Stack

• Backend: Python 3.10, Flask
• AI Models: GPT-5 via OpenAI API
• Database: SQLite or PostgreSQL (depending on your setup)
• Containerization: Docker, Docker Compose, Docker Swarm
• Frontend: HTML/CSS/JS (templates in UI2/)

---

Getting Started step by step

1. Clone the repository

git clone https://github.com/brsynth/BioBot.git
cd BioBot

2. Activate Docker Swarm (for secrets)

sudo docker swarm init

3. Create the required Docker secrets

You need an OpenAI API key to use Biobot.

echo "YOUR_OPENAI_API_KEY" | sudo docker secret create biobot_api_key -
echo "YOUR_DB_PASSWORD" | sudo docker secret create db_password -

Check the secrets :

sudo docker secret ls

4. Build the image

sudo docker build -t biobot:latest ./UI2

5. Deploy the stack with docker swarm

sudo docker stack deploy -c docker-compose.yml biobot

Check that the service is running:

sudo docker service logs -f biobot_biobot

6. Access the app

The app will be accessible via

http://127.0.0.1:5000

Usage

• Register as a new user or log in with an existing account.

• Start a conversation in the chat interface.

• Ask general lab questions or request code generation for lab protocols.

• The system will classify your request, fetch documentation, and generate Python code if needed.

• All chats are stored in the database and used to provide BioBot with memory.