In the digital age, the rapid spread of misinformation presents a critical challenge, necessitating advanced, scalable solutions for accurate and efficient fact-checking. Addressing the gap in Singapore-specific multimodal misinformation detection and overcoming limitations in current automated models, this project introduces SINDEX, which has achieved strong performance across all functions, and state-of-the-art performance in the verification of Singapore-specific multimodal misinformation. Our automated evidence retrieval framework, which leverages the H2-keywordextractor transformer model and SBERT to query NewsAPI, ensures greater accuracy practicality. We developed a CLIP multimodal Custom Truthfulness Classifier trained on our self-collected dataset of Singapore-context multimodal misinformation and a parallel ChatGPT-4-32k-involved misinformation detection system, achieving robust F1-scores of 95.2% and 95.4%, respectively. Additionally, a custom SBERT-based satire detector was implemented to assess exaggerated content, yielding a strong F1-score of 93.2%. For out-of-context misinformation, we verified that the Catching Out-of-Context Misinformation using Self-Supervised Learning (COSMOS) model was suitable for the domain transfer task and executed custom code for its implementation. Lastly, we developed a system for explanation generation by prompting ChatGPT-4-32k with optimised input parameters which has achieved high average BLEU, METEOR, and ROUGE scores of 84.1%, 84.5%, and 91.3% respectively. SINDEX promotes an informed digital environment, empowering users to make evidence-based decisions, thus enhancing digital literacy and resilience against misinformation. Designed for scalability, SINDEX is readily adaptable to global misinformation challenges.