The current vehicle insurance claim assessment process suffers from inefficiencies due to manual inspections, resulting in delays, inaccuracies, and high operational costs. Existing AI-based solutions face challenges with real-time scalability, dataset limitations, and limited adaptability to complex damage scenarios. Furthermore, the absence of MultiAgentic Systems often leads to suboptimal repair estimations, while ineffective classification models contribute to high false positive rates. To address these challenges, we propose Insurify, a hybrid AI-powered framework that integrates Agentic Systems with classical machine learning and deep learning models for real-time, accurate vehicle damage assessment. Our approach leverages stable vision-language models, graph-based damage representation, and continuous learning mechanisms to enhance classification accuracy and reduce false positives. By utilizing modular AI pipelines, real-time WebCrawling Agents, and Knowledge Graphs, Insurify offers a scalable and adaptive solution for insurance companies. Experimental evaluations demonstrate significant improvements in damage assessment accuracy, cost estimation reliability, and processing efficiency. This work pioneers a comprehensive AI-driven insurance automation framework, bridging critical gaps in existing methodologies and setting new standards for real-time claim assessments.