CertiK launches Skill Scanner to establish a standardized security review layer for AI Agent applications

Foresight News reports that security company CertiK has announced the launch of its AI Agent ecosystem security product, "CertiK Skill Scanner." The product is suitable for both the Web3 ecosystem and the traditional Web2 market, targeting the AI Skill marketplace, enterprises, and developers. It is designed to identify risks such as malicious behavior, data leaks, privilege escalation, and autonomous execution before Skills are executed, achieving a risk detection accuracy rate of 90.5%.

Unlike traditional AI scanning tools, CertiK Skill Scanner supports not only static code analysis but also evaluates the risks during actual Skill operation. It is particularly suitable for scenarios involving financial transactions and fund calls. Currently, the tool has already been integrated into AI Agent platforms such as Pieverse.