Augmenting API Security Testing with Automated LLM-Driven Test Generation

In this paper, we propose a novel approach that leverages Large Language Models (LLMs) to automatically generate Karate DSL test cases from OpenAPI Specification documents. Our approach aims to exploit the natural language understanding and generation capabilities of LLMs to produce high-quality and comprehensive test cases that cover various security aspects. We present a proof-of-concept framework that integrates LLMs with Karate and demonstrates its feasibility and effectiveness on a sample API. We also discuss the limitations and future directions of our approach, which opens up new possibilities for applying LLMs to API security testing and other complex domains.