AI MCP Management
Extend the Emporix Agentic AI capabilities with external MCP servers.
Emporix comes with a set of built-in MCP servers that are ready for your usage out-of-the-box, without any further configuration. The default MCPs cover the common ecommerce operations and are fully integrated with the Emporix Commerce and Orchestration Engine. However, we realize that you might require more flexibility as many businesses rely on additional specialized systems, such as for example ERPs, CRMs, or other custom services. That's why custom MCP servers allow the custom agents to securely interact with these external systems, enabling them to go beyond what the default ecosystem provides.
For instance, imagine that you need an agent that can communicate directly with your ERP system to fetch product availability or to process purchase orders. With the AI MCP Management, such complex interactions become possible and straightforward. Enabling custom MCP servers empowers you to design bespoke agents and integrate them seamlessly with your broader IT landscape, tailoring their functions to your exact business processes.
Configuring custom MCP server
Choose to add a custom MCP server
In the Management Dashboard, go to the Agentic AI -> AI MCP and choose Add new MCP server.
Define the MCP server
Provide the configuration details for your custom MCP server.
MCP Server ID – A unique identifier for the external MCP configuration that the system is to reference your server by.
MCP Server Name – A human-readable name for the MCP server (for example, ERP Connector).
Transport – The transport layer used for communication. Currently, the Server-Sent Events (SSE) type is supported.
URL – The endpoint URL of your external MCP server.
Authorization Header Name (Optional) – If your MCP server requires authorization, specify the header name to be included in requests (for example, Authorization).
Authorization Header Token ID (Optional) – Select a token (defined previously in the AI Tokens view) to be used as the value of the authorization header.
This allows secure, reusable credential handling without exposing sensitive data.

The MCP Server is enabled in the system and you can use it now in your agents.
Security considerations
Tokens used for authorization are never displayed once saved — they remain encrypted and securely managed.
Only the selected Authorization Header Token ID is passed to the server, ensuring credentials stay isolated and controlled.
The centralization of tokens and MCP configurations ensures consistency and auditability across all integrations.
In essence, a custom MCP server is crucial for businesses that need their AI agents to interact with a unique set of external systems, enabling highly tailored automation and deeper integration across their entire workspace.
Last updated
Was this helpful?

