In the landscape of modern software development, security threats are pervasive; therefore, it is essential that security risk management and vulnerability checks be performed throughout development. One of the common approaches to bug finding and security vulnerabilities is source code static analysis. However, it tends to be a time-consuming process, prone to errors and not well-organized. Upon the detection of the vulnerabilities, these need to be repaired, and hence the cycle goes on for detection and patching. In this paper, it proposes an efficient pipeline from source code vulnerability detection and restoration for efficiency and accuracy improvement with an automated approach. We used the combination of our custom solution to integrate automated mechanisms at each stage of the scanning process for reducing the scope of human manual intervention and minimizing the increase of false positives and hastening the pace of a secured resolution with the support of multiple AI agents. A close approximation of the proposed framework gives us a view which has possible ways to reduce security overhead at code reviews without necessarily exposing the integrity and security of software by using multiple AI Agents (LLMs), as means for appropriately detecting and repairing a host of vulnerabilities in source codes.