Sherlock Launches New AI Audit Engine to Improve Code Security

Sherlock has officially launched its Audit Engine, a platform designed to streamline how blockchain protocols identify security vulnerabilities. The tool works by coordinating multiple AI systems and human researchers to review the same codebase simultaneously. This process allows Sherlock to collect findings from different sources, validate the data, and provide a single, comprehensive report.

The platform recently proved its effectiveness through a high profile test with Polygon. Earlier this year, Sherlock used the system to audit the Heimdall V2 consensus client. The project showed that no single AI model catches every bug, confirming the value of combining various automated tools with human oversight. By running these checks together, the engine captures a much larger picture of potential threats.

This launch comes at a time when cybersecurity AI is advancing rapidly, with new models like Gemini 3.5 Flash Cyber entering the market. The Audit Engine is built to adapt as these technologies improve. It offers protocols a consistent way to measure which tools are actually working well for their specific code needs.

For the crypto industry, this means security audits could become faster and more reliable. As new models arrive, Sherlock plans to integrate them into its platform, giving developers a clearer view of their project's safety. Observers are watching to see how quickly major protocols adopt this multi layered approach to protecting their code.