Skip to main content

Creating MCP

MCP Usage and Creation Guide​

This chapter introduces how to create, configure, and use MCP (Model Control Plane) in SERVICEME's AI Studio.
With MCP, you can introduce external capabilities to Agents, test tool availability, publish personal Agents as organization-level MCPs, and invoke these functional plugins in real business scenarios.

Introducing External MCP​

Operation Path:

  1. Click “MCP” in the left navigation bar of the AI Studio page to enter the MCP management page;
  2. Click the “Create” button at the top right corner of the page to enter the MCP creation interface;
  3. Fill in the basic information and complete the configuration.

Creation Content Description:

  • MCP Name: Set a unique name for the MCP to identify it.
  • MCP Avatar: You can choose the default avatar provided by the system (custom uploads are not supported yet).
  • MCP Description: Briefly describe the MCP’s functions, application scope, and purpose.
  • MCP Category: Select the category to which the MCP belongs for easier management and searching later.
  • MCP Service Configuration:
    Register external services (such as APIs or system interfaces) by filling in JSON format code, thereby adding external invocation capabilities to the assistant.
    For example, you can connect services like weather, translation, reporting systems, etc., enabling the Agent to have richer functions.

Tip:
Be sure to save after configuration; otherwise, the MCP will not take effect during testing or invocation.

Testing MCP Tools​

After completing the MCP service configuration, you can verify the availability of the MCP through the built-in testing function:

  1. Click the “Test” button on the MCP configuration page;
  2. The system will automatically execute the MCP Tools test;
  3. After the test is completed, the “MCP Tools” area on the right will display the list of available tools provided by the MCP.

Testing Suggestions:

  • It is recommended to test each tool item by item to ensure the interface responds normally;
  • If the test fails, please check whether the JSON configuration is correct and whether the interface is accessible.

Publishing an Agent as MCP​

Besides introducing external resources, you can also publish your created Agent as an MCP, allowing other members within the organization to use the Agent’s functions.

Operation Steps:

  1. Create a personal Agent and complete the basic configuration;
  2. Enter the configuration page of the Agent;
  3. Select “Publish as MCP” in the settings;
  4. After confirming the publication, the system will convert the Agent into an organization-level MCP.

Application Description:
The published MCP can be directly referenced by other members within the organization in their own Agents,
which is very suitable for sharing mature intelligent capabilities (such as automated reporting, knowledge retrieval, process approval) with the entire team.

Finding and Using MCP​

In an already created Agent, you can select and enable existing MCP services for it.

Usage Steps:

  1. In the detailed configuration page of the Agent, find the “MCP Services” area;
  2. According to business needs, select and enable the corresponding MCP from the list (such as the add function in the calculator plugin);
  3. After adding, click “Publish” at the top right.

After successful publishing, the Agent can use the capabilities provided by the selected MCP.
For example:

  • Enter “1+1=” in the chat interface,
  • The Agent will perform the calculation through the calculator MCP and display the answer with an MCP invocation mark.

Practical Suggestions​

  • Quick Start: For business users, it is recommended to first try the published organization-level MCPs, which can be called directly without manual configuration.
  • Flexible Expansion: Users with some technical background can connect external systems via JSON to expand enterprise-level AI capabilities.
  • Collaboration and Sharing: Publishing frequently used Agents as MCPs can improve team reuse efficiency.

By following the above steps, you can flexibly introduce, test, share, and use MCP in SERVICEME,
making intelligent agents truly integrated into your business systems and enhancing the practicality and scalability of AI automation.