Creating friendly, welcoming spaces is a common consideration when building collaborative applications. To empower end users to better deal with the issue of manually moderating their sessions, developers can build applications that use automated text chat moderation options. This post explores one such option using TensorFlow.js in Daily-powered video applications by extending the chat interface of Daily's Vue demo app. The approach taken is just one exploratory option for aided moderation and does not aim to be foolproof, as human interaction is hard. The implementation involves installing the TensorFlow.js package, loading the toxicity model, and analyzing input text using the model. When toxic content is detected, a warning message is displayed to the user. This demo serves as an example of how to implement client-side chat moderation capabilities in Daily applications, but it's not intended for production use due to its experimental nature and potential risks of bypassing checks by malicious actors.