Fake news has been an ever-increasing issue to be addressed due to advancements in using social media platforms, as with the ease of information sharing, it has become increasingly challenging to ensure the dissemination of reliable information. While numerous researchers have proposed detection methods, these approaches often fall short against sophisticated adversarial attacks which target the vulnerabilities of AI models. TextAttack, a Python library, is one such method to do adversarial attacks but in contrast, our paper implements TextAttack as a defense strategy by leveraging its built-in data augmenter recipes. We conduct a comprehensive evaluation to assess the resilience of fake news detection against adversarial manipulations. Our research emphasizes on the capabilities of TextAttack in identifying vulnerabilities and fortifying defenses against adversarial attacks. By integrating TextAttack's data augmentations to safeguard against adversarial attacks, significant advancement in the development of reliable and robust fake news detection systems with better generalization to variations is proposed, addressing the critical issue of misleading information on digital communication platforms.