This repository contains the evaluation framework for the paper "Prompt Engineering Trumps Tool Selection: A 540-Test Study of AI-Generated Infrastructure Code Security", comparing the security of Infrastructure-as-Code generated by GitHub Copilot, Cursor, and Windsurf.
@article{buck2025prompt,
title={Prompt Engineering Trumps Tool Selection: A 540-Test Study of AI-Generated Infrastructure Code Security},
author={Buck, Lukas},
year={2025},
url={https://www.researchgate.net/publication/397401473_Prompt_Engineering_Trumps_Tool_Selection_A_540-Test_Study_of_AI-Generated_Infrastructure_Code_Security}
}