OX Security Discloses Systemic RCE Baked Into Anthropic's Official MCP SDKs Across Python, TypeScript, Java, and Rust

OX Security·high signal

OX Security's research team reports a critical, architectural vulnerability in the official Model Context Protocol SDKs that enables arbitrary command execution (RCE) on any system running a vulnerable MCP implementation, exposing internal databases, API keys, and chat histories. Crucially they frame it as a design decision present across every supported language SDK — Python, TypeScript, Java, and Rust — not a single-language bug. This is distinct from the back-end-connector and unauthenticated-server classes already reported, and warrants immediate review of any self-hosted MCP server.