Create an MCP tool

MCP (Model Context Protocol) tool is used to integrate multi-cloud platform services or run local client applications.

Quickly create MCP tool

1. Go to the tool management interface

First, please enter the " AI Features" section, then click "🔧 Tools". After entering the tool list page, click the "➕ Add Tool" button in the top right corner.

2. Choose tool type

Please choose tool type MCP.

3. Set display name

Set a clear display name for the tool, here set as Composio mcp for google calendar.

• Purpose: This name will be shown in the platform interface for all users to see.

• Recommendation: Choose a name that clearly expresses the main function of the tool to help users understand. There is no strict format requirement for this name.

4. Fill in MCP configuration

a. 🔗MCP server URL

• Purpose：MaiAgent currently accepts external MCP servers. By providing the service address (URL) of an MCP server, the AI can call MCP services to connect external applications.

• Format:

  • Please enter the full URL (for example:https://mcp.dev/maiagent/mcp_service).

• Note：This field is required.

Paste your MCP server URL here; the system will automatically fetch the list of tools linked in that server:

b. 🎛️ MCP command arguments (mcp_args)

• Purpose：Defines the names of parameters that need to be passed when executing MCP commands or calling MCP services; the content is automatically generated by the AI.

• Format：It is recommended to use JSON array format, where each element is a string representing a parameter.

  • Examples (JSON array):

    Copy

    [
        "--user",
        "admin",
        "--config",
        "/path/to/config.yaml"
    ]

  • At runtime：The AI will pass these parameters to the MCP tool in order

  • If you enter a comma-separated string (for example:arg1,arg2,arg3), the system will try to parse it as a list of arguments. To avoid ambiguity, using a JSON array is recommended.

c. 🌳 MCP environment variables (mcp_env)

• Purpose：Set the necessary environment variables for the execution environment of MCP commands.

• Format：Must be a valid JSON object where the key is the environment variable name and the value is the environment variable content (string).

  • Examples:

    Copy

    {
      "API_KEY": "{{SECRET_MCP_API_KEY}}",
      "REGION": "us-west-1",
      "DEBUG_MODE": "true"
    } 

5. Find “Allowed tools (JSON array)”, then tap Refresh

• Purpose：Specifies the concrete list of sub-tools that the AI is authorized to use under this MCP client. A single MCP client may provide multiple different functions or sub-tools.

• Auto-detect/Leave blank：If this field is left blank or not provided, when the system first connects to the MCP client it will attempt to automatically detect all available sub-tools and by default allow all detected sub-tools. If you want to restrict the AI to only use specific sub-tools, please list them explicitly here.

After tapping, the system will automatically fetch the tool contents linked to that server and display them in the list:

6. 💾 Save tool

After confirming all settings are correct, scroll to the bottom of the page and click the "Confirm" button. Your new tool is now created!

⚠️ Important reminder

Connection test

• After creating the tool, it is recommended to first test whether the MCP connection is working properly

• You can verify the tool functionality in a test environment

Permission management

• Carefully choose the allowed tools to avoid granting too many unnecessary permissions

• Regularly check tool usage

Troubleshooting

• If the connection fails, please check whether the MCP server URL is correct

• Confirm that environment variables and parameter formats meet the requirements