Control Your Screens from Claude or ChatGPT

Most digital signage software gives you a dashboard. Brix gives you a dashboard and an interface an AI assistant can drive directly, so you can ask Claude or ChatGPT to change what is on your screens and it does it.

This is not a chatbot bolted onto a settings page. Brix runs a Model Context Protocol server: a standard way for an AI assistant to discover what a product can do and then do it. Connect it once and the assistant can see your screens, check why one went offline, swap a playlist, or pull a proof-of-play report.

What you can actually ask for

The connection exposes 42 tools. A few of the ones people reach for first:

• "Which screens are offline, and why?" The assistant reads the outage history for each screen and tells you the cause, not just the state.
• "Put the winter menu on every screen in the Leeds shops." It finds the screens by location and assigns the content.
• "How many times did the safety video play last month?" It runs a proof-of-play report over the period you name.
• "Something is wrong with the screen in reception." It runs a diagnosis and reports what it found.

You do not learn a command syntax. You describe the outcome.

Why this matters more than it sounds

The usual reason to have an API is that a developer will write something against it one day. The reason to have an MCP server is that the person who runs the screens can act on them today, in the tool they already have open, without learning your dashboard.

That is a real difference for a business with more screens than staff. The manager who needs tomorrow's specials on eleven screens does not want a training course. They want to ask.

Set it up

You need an API key. Open Settings, then API and MCP, and select New key.

Give the key a name you will recognise later, choose which locations it can reach, and pick its permissions. A key can be full access, or it can be limited to exactly the actions you want an assistant to take.

Scope the key to the smallest set of permissions that does the job. If you only want an assistant to answer questions about your screens, give it read permissions and nothing else. It will still be able to tell you which screen is down and why. It will not be able to change what plays.

The full walkthrough, including the connection details for each assistant, is in API keys and integrations.

Every tool says what it will do

Each of the 42 tools declares whether it only reads, whether it changes something that already exists, and whether it reaches anything outside your workspace. Assistants use those declarations to decide when to check with you first.

So an assistant will answer "which screens are offline" without stopping to ask. It will pause before taking over every screen with an emergency message, because that tool is marked as changing something that exists. The permission you set on the key is the hard limit; these declarations are what make the assistant behave sensibly inside it.

What it costs

Nothing extra. The API and the MCP server are included on every Brix plan, at $6 per screen per month. There is no integrations tier and no per-call charge.

That is worth saying plainly, because it is unusual. On most platforms the API is the thing that moves you up a pricing tier. See what the alternatives charge for API access.

Start here

If you already run Brix, open Settings, then API and MCP, and make a read-only key. Ask your assistant which screens are offline. It takes about two minutes and you will immediately see whether this fits how you work.

If you do not run Brix yet, the free trial includes it.