TestFit Model Context Protocol (MCP) - FAQ
TestFit MCP is currently only available for the desktop app. Web functionality is not available at this time.
What is the Model Context Protocol (MCP) connector?
It is a way to drive TestFit with the AI assistant you already use. Your assistant sends requests
to TestFit, and the algorithmic engine runs inside the TestFit software. You get to prompt and
generate work in TestFit through a tool you are already comfortable with.
Do I bring my own AI, or does TestFit provide one?
You bring your own. You choose which assistant to connect to, whether that is Claude, ChatGPT, Microsoft Copilot, or another provider. TestFit does not supply, embed, or recommend
a model. The connection simply lets your assistant talk to TestFit.
Where does the AI actually run?
On your side. The model and everything that powers it stay with you and your provider. TestFit's
role is to receive requests and return output. The intelligence is yours; the engine is ours.
What does TestFit send back?
TestFit returns the same kind of output you get in the product today: site configurations,
geometry, and analytics generated by TestFit's proprietary algorithms. There is no separate AI
inside TestFit interpreting your request. The engine is deterministic, which means the same
inputs produce the same reliable results.
What can I do with it?
Anything the connector exposes is expressed in plain language through your assistant. You can
ask your assistant to set up and generate site plans, adjust inputs, and pull results without
leaving the tool you prompt from.
Do I need to change how I use TestFit?
No. The connector is an additional way to work. Everything you do in TestFit directly continues
to work exactly as before.