Recently, I paid attention to some open-source projects and technical articles related to security that involve large language models (LLMs). The authors emphasized the use of LLMs, but a closer look…