Model Context Protocol (MCP)
 |
The Model Context Protocol (beta) is currently offered in preview mode and provides only basic functional validation. Full qualification and expanded capabilities will be introduced in future releases.
|
The Model Context Protocol (MCP) is a standardized protocol that enables artificial intelligence systems, especially large language models (LLMs), to securely and intelligently interact with applications.
MCP adds structured context awareness to the ThingWorx Platform by exposing executable units of work (such as ThingWorx Services) as standardized tools, prompts, and resources that LLMs can understand and invoke. This approach helps LLMs to make accurate, context-driven decisions rather than executing raw service calls without awareness.
Because MCP defines a universal contract, ThingWorx capabilities can be easily integrated into MCP-compatible environments such as Claude Desktop, Visual Studio Code, and other supported clients.
The following sections build on this foundation by exploring key concepts, security considerations, practical examples, and the core components that make MCP work in ThingWorx.