Articles on: AI & MCP

What is MCP and How Does It Work with COUNT?

MCP (Model Context Protocol) is what allows AI assistants like Claude and ChatGPT to securely connect to your COUNT workspace.

MCP acts as the connection layer between COUNT and AI tools, allowing them to retrieve information and perform approved actions in your workspace.

What is MCP?

MCP stands for Model Context Protocol.

Think of MCP as a secure connection between COUNT and AI assistants.

Without MCP, AI assistants cannot access your COUNT workspace or interact with accounting data.

With MCP, connected AI assistants can securely retrieve information and perform approved actions in COUNT.

COUNT uses MCP to allow AI assistants to work with accounting data in a secure, structured way.

Instead of manually searching for information, you can ask questions naturally and let AI retrieve the relevant data.

For example:

“List all of my transactions from COUNT last month”

Or:

“Show overdue invoices over $1,000”

MCP allows COUNT to understand what information is needed and securely provide it to the connected AI assistant.

When you ask an AI assistant a question that requires COUNT data, the assistant can request access to the relevant COUNT tool.

Here’s how the process works:

Step	What happens
1	You ask a question in Claude or ChatGPT
2	The assistant detects it needs COUNT data
3	COUNT requests permission
4	You approve the request
5	The assistant retrieves and formats the results

For example, if you ask:

“Create a customer with a [customer name]”

You may see a request such as:

“Claude wants to use Create Customer from COUNT”

After approval, the assistant retrieves the information and presents it in an easy-to-read format.

When an AI assistant needs access to COUNT data or wants to perform an action, COUNT may ask for permission.

You can choose to:

This helps ensure you remain in control of what AI assistants can access or do in your workspace.

Updated on: 26/05/2026

Was this article helpful?

Thank you!