A large language model-based analysis of vulnerability discovery in windows software

Source code security auditing is essential before software release in order to identify programming faults that may lead to vulnerabilities and functional failures. In this paper, we present a structured security assessment of the Windows App SDK by integrating multiple static analysis tools with a context-aware and disagreement-aware Large Language Model (LLM) interpretation layer. Although static analyzers are effective in reporting potential weaknesses, their raw outputs often contain redundant alerts, limited contextual explanation, and inconsistent severity assignments. To address these limitations, the proposed LLM-based interpretation layer normalizes and de-duplicates alerts, filters context-limited or nonactionable warnings, and refines severity prioritization under inter-tool disagreement without introducing new vulnerability discoveries. Experimental results show the security findings before and after LLM-based interpretation. In particular, the proposed framework reduces static-analysis alerts by 62.5%. In addition, disagreement-aware severity refinement eliminates over-prioritized critical findings and improves prioritization by reducing Medium findings from 11 to 7 and Low findings from 42 to 28. These results demonstrate the potential of LLM-based interpretation to reduce noise in static-analysis outputs and improve vulnerability prioritization for practical security assessment.