[Project in Progress ... ]

Testing System for AI and Automation Tools

This project focuses on comparing, evaluating and analyzing the performance of traditional testing tools, intelligent AI agents, and Large Action Models (LAMs). The system provides a unified dashboard for monitoring and managing the testing process, using a prototype hospital management application, CuraNava, as a test bed.

CuraNava serves as a modular and scalable application that includes essential hospital management features such as patient management, appointment scheduling, medical test result management, and medicine inventory. Although future plans may involve implementing three quality levels for each feature (poor, medium, high), these levels are not currently available.

Table of Contents

• Testing System Features
• System Architecture
• Project Structure
• Setup and Deployment
• Evaluation Methodologies
• Contributing
• License

---

Testing System Features

General

• Unified dashboard for managing tests and visualizing results
• Integration with traditional testing tools, AI agents, and LAMs
• Modular architecture to support multiple application types
• Exportable reports for test analysis

CuraNava Prototype Features

• Patient Management: CRUD operations for patient profiles, advanced search, and audit logging
• Appointment Scheduling: Calendar views, automated reminders, and conflict detection
• Medical Test Results: Secure uploads, categorization, and notifications
• Medicine Inventory: Stock tracking, low-stock alerts, and reporting
• Security: Role-based access control and two-factor authentication

Version Switching for Future Testing

• Planned feature: Toggle between poor, medium, and high-quality versions of each functionality to assess testing tool adaptability

---

Prototype System Architecture

Frontend

• Technology: React
• Features:

Backend

• Technology: Node.js with Express
• Features:

Database

• Type: SQLite database
• Features:

Deployment

• Platform: (tba)
• Version Control: Git (branch management, regular updates)

---

Project Structure

BachelorThesis_IFI/
├── curanava/                   # The Prototype App with Features to be Tested
├── testing_tools/              # The Testing Tools Applied to the Prototype
├── management/                 # Integration with services like JIRA, OpenAI, GitLab ...
├── assesment/                  # Assesment Tools and Scripts for comparing the different Testing Tools
├── shared_configs/             # Project Configurations (like database configs etc)
├── dashboard/                  # Visual Interface of the Testing Workflow
├── docs/                       # Documentation files and guides
└── README.md                   # Project overview

---

Setup and Deployment

Prerequisites

• React with Typescript via Vite
• Node.js
• SQLite database (later MySQL, PostgreSQL)
• Git

Installation (Windows 10, 11)

(In progress ...)

1. Clone the repository:

   git clone <repository-url>
   

2. Install dependencies:

   
   # Installing Web App to be tested
   cd BachelorThesis_IFI/0_curanava/
   npm install
   
   cd frontend
   npm install
   cd ../backend
   npm install
   
   # Installing playwright
   cd ../../1_testing_tools/01_classic_automated_testing/playwright
   npm init playwright@latest (choose TypeScript, e2e tests: tests/e2e, Install Playwright browsers true)
   
   # Installing Testzeus Hercules
   cd ../../02_agentic_testing/testzeus_hercules
   python -m venv hercules-venv
   ./hercules-venv/Scripts/activate
   pip install testzeus-hercules
   playwright install --with-deps
   mkdir input
   mkdir output
   mkdir test_data
   
   # Installing xLAM required dependencies
   cd../../03_lam_testing/xlam
   python -m venv xlam-venv
   ./xlam-venv/Scripts/activate
   pip install transformers
   pip install torch
   pip install datasets
   pip install tokenizers
   pip install 'accelerate>=0.26.0'
   (pip install -r requirements.txt)
   
   
   

3. Set up the database:

   • Configure the database connection in .env

4. Start the development server:

   (in progress)
   

5. Make a copy of .env-example, rename it to .env and enter your credential for the different services

Deployment

• (in progress)

---

Evaluation Methodologies

This system supports the evaluation of:

1. Traditional Automation Tools: Tools like Playwright for automated UI and functional testing.
2. AI Agents: Agents such as TestZeus Hercules for adaptive and intelligent test execution.
3. Large Action Models (LAMs): Models like xLAM for dynamic and context-aware testing workflows.

---

License

This project is licensed under the MIT License. See LICENSE for details.