This security guide maps vulnerability classes in Model Context Protocol deployments, including prompt injection, command injection, and token-related attacks. It covers control strategies such as authentication enforcement, capability scoping, and least-privilege access at the execution layer.