Recently, social media and online communication tools have become an integral part of digital technology, and the widespread use of the same has definitely increased the rate of digital interactions. However, it has also increased the rate of cases of cyberbullying and other types of abusive online communications. Online communications, such as abusive words, harassment, and also the use of slangs and other sorts of abusive words that reflect upon the psychological state of individuals, are on the rise and are having a major impact on individuals. Traditional mechanisms of identifying online abusive communications are either not effective enough to work efficiently with a large database of communications or are not aware of the contexts. In this regard, this paper proposes a framework of a web-based cyberbullying detection tool known as CyberGuardian that attempts to identify abusive communications submitted to the system by the users. In this paper, we proposed the development of a tool that aids in checking user-submitted communications to identify abusive words and context. The tool has been implemented with a modular Flask framework and can be said to have low computational complexities, and the framework has been highly effective in identifying explicit abusive words