Clear and effective writing is essential for successful communication in diverse personal, professional, and academic contexts. However, crafting high-quality text can be challenging, requiring proficiency in grammar, nuanced expression of sentiment, and accurate conveyance of emotion. While existing writing assistance tools offer some support, they often need more holistically addressing these multifaceted aspects of writing. This research presents a novel Natural Language Processing (NLP) pipeline designed to provide comprehensive writing assistance by integrating four key functionalities: grammatical acceptability classification, grammar correction, sentiment analysis, and emotion detection. We leverage the power of fine-tuned pre-trained transformer models, specifically RoBERTa and FLAN-T5, to achieve robust performance across these tasks. Our pipeline employs a modular architecture, allowing for specialized training and evaluation of each component.