Intelligent Test Automation: A Multi-Agent LLM Framework for Dynamic Test Case Generation and Validation
Pragati Kumari - · International Journal on Science and Technology · 2025
Automated software testing is essential in modern software development, ensuring stability and resilience. This study describes a unique technique for using the capabilities of Large Language Models (LLMs) via a system of autonomous agents. These agents collaborate to dynamically generate, validate, and execute test cases based on specified requirements [1, 2]. By iteratively improving test cases via agent-to-agent communication, the system improves accuracy and effectiveness. Our implementation, which uses AutoGen and Python's unittest framework, shows how this method helps to maintain excellent software quality. Experimental evaluations across a variety of test scenarios demonstrate the versatility and efficiency of our framework, Intelligent Test Automation (ITA), emphasizing its promise for increasing automated software testing [3, 4].