MCP-Scanner: Detecting Security Risks in Model Context Protocol Systems

The Model Context Protocol (MCP) was introduced to give AI models a unified way to interact with external tools and services, but its rapid adoption has outpaced the development of its security model — MCP was released with a flexible, underspecified design that leaves its open architecture exposed to new classes of security risk. MCP-Scanner is introduced as the first comprehensive tool designed to automatically detect a wide range of vulnerabilities in MCP servers, giving developers and operators a way to audit MCP deployments before they are exposed to production AI agents.