AI Vulnerability Verification
Supports providing issue audit suggestions to users through AI models. By collecting data from code repositories, historical issue data, false positive cases, and correct examples from a knowledge base, and then preprocessing and labeling this data, the control flow and data flow information of the code (such as function call relationships, dependency relationships, library function documentation, etc.) are used to build a knowledge graph. This knowledge graph is integrated into the model, enabling it to achieve the capabilities of a code security audit expert. The model can comprehensively understand the context of the code and the relationships between functions, providing accurate detection results and audit reasons.
This capability helps software development teams quickly and accurately obtain detection results and audit reasons comparable to those provided by security experts during the code audit process. Especially when dealing with large-scale codebases, the AI-powered false positive detection feature of Xmaze AI significantly reduces the workload for security auditors, thereby improving review efficiency.
Reduce audit time by 90%.