The increasing demand for accessible mental health care has driven innovation in artificial intelligence (AI) technologies for therapeutic assistance. This paper introduces SoulCare, a locally-executed AI chatbot designed to provide mental health assistance. By fine-tuning a Large Language Model (LLM), SoulCare is tailored to understand and respond effectively to various mental health-related questions. In contrast to cloud-based systems, SoulCare runs securely on the user's local system, providing greater privacy and data protection. The chatbot is designed for high-performance efficiency with a user-friendly interface, to enhance access to mental health services. This paper illustrates how cutting-edge AI technologies can be repurposed for sensitive applications, such as mental health, without compromising security and usability.