Elgato added MCP support to Stream Deck: AI now presses buttons

Elgato released the Stream Deck 7.4 update, which adds support for Model Context Protocol — an open standard for AI agents to interact with external tools. Now Claude, ChatGPT, and Nvidia G-Assist can press buttons on the device instead of the user.

The working principle is simple: the user still configures actions in the Stream Deck application in the usual way. MCP adds a new layer of control on top of the existing configuration. After connecting an AI assistant to the device, it's enough to write or say a request — for example, "start streaming" or "switch scene" — and the agent will find the right action and activate it.

Stream Deck has long ceased to be a device only for streamers. It is used by video editors, designers, developers, and anyone who works with many repetitive operations. Physical buttons with custom icons allow you to automate routine tasks with one press — from microphone management to running scripts and switching between applications.

MCP support logically continues this idea: now an intermediate element can be not a finger, but a voice command or a text request to a chatbot. Elgato is not the first company to integrate MCP into consumer hardware and software. Since the open specification of the protocol was published in late 2024, it has been adopted by dozens of platforms — from IDEs and browsers to enterprise tools.

Stream Deck 7.4 is one of the first examples of MCP in devices for content creators and advanced users. What this means in practice: the boundary between manual control and automation becomes even thinner.

Users who have already built complex profiles in Stream Deck gain voice and text interface to them for free — without rewriting the configuration. For others, this is a reason to finally figure out the device: if AI can explain and launch actions, the entry barrier is significantly lowered.