Create MCP Tools

This guide will walk you through creating new MCP tools on the platform.

What is MCP?

MCP, which stands for Model Context Protocol, is used to integrate multi-cloud platform services or execute local client applications.

Tool usage can be imagined as a power plug - services can only be used after plugging in. Traditionally, each LLM develops different ways of using tools, like different plug specifications requiring multiple sockets. If you want both OpenAI and Claude to use Google Calendar services, developers would need to design multiple scripts for different AIs to use.

When Google Calendar updates, developers may need to:

1. Update 3 different versions of integrations simultaneously

2. Maintain 3 different technical documents

This leads to lengthy development timelines and high costs.

MCP Solution

MCP is a standardized protocol designed to solve this problem through:

1. Standardized tool definition format

2. Unified communication protocol

3. Consistent error handling mechanism

The above processing simplifies the development workflow.

With MCP, developing a tool only requires:

1. Building MCP Server (one-time development)

2. Defining standardized tool specifications

3. Implementing unified business logic

4. All AI platforms supporting MCP can use it

Differences before and after MCP integration

MCP is essentially a protocol designed to standardize AI tool integration, allowing development teams to "develop once, use everywhere", significantly reducing the complexity and maintenance costs of the AI assistant tool ecosystem.

Quick Creation of MCP Tools

1. Access Tool Management Interface

First, navigate to the "AI Features" section from the left navigation bar, then click "🔧 Tools". After entering the tools list page, click the "➕ Add Tool" button in the top right corner.

Click "➕ Add Tool" to begin creation

2. Select Tool Type

Choose MCP as the tool type.

3. Set Display Name

Set a clear display name for the tool, in this case Composio mcp for google calendar.

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

• Recommendation: Choose a name that clearly expresses the tool's main function for easy user understanding. There are no strict format restrictions for this name.

4. Fill in MCP Configuration

a. 🔗MCP Server URL

• Purpose: MaiAgent currently accepts external MCP servers. By providing the MCP server's service URL, AI assistants can call MCP services to connect to external applications.

• Format:

  • Enter the complete URL (e.g., https://mcp.dev/maiagent/mcp_service).

• Note: This field is required.

For information on obtaining MCP URLs, refer to Technical Staff Manual—Remote MCP Service Overview

Paste your MCP server URL here, and the system will automatically fetch the list of tools connected to that server:

b. 🎛️ MCP Command Parameters (mcp_args)

If no specific environment variables need to be set, this field can be left empty.

• Purpose: Define parameter names to be passed when executing MCP commands or calling MCP services, with content automatically generated by AI assistants.

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

  • Example (JSON array):

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

  • During execution: AI assistants will pass these parameters to the MCP tool in sequence

  • If you input a comma-separated string (e.g., arg1,arg2,arg3), the system will attempt to parse it as a parameter list. To avoid ambiguity, using JSON array is recommended.

Set MCP command parameters

c. 🌳 MCP Environment Variables (mcp_env)

If no specific environment variables need to be set, this field can be left empty.

• Purpose: Set necessary environment variables for MCP command execution environment.

• Format: Must be a valid JSON object where keys are environment variable names and values are environment variable contents (strings).

  • Example:

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

Set MCP environment variables

5. Find "Allowed Tools (JSON Array)" and Click Refresh

• Purpose: Specify the list of specific sub-tools that AI assistants are authorized to use under this MCP client. An MCP client may provide multiple different features or sub-tools.

• Auto-detect/Empty: If this field is left empty or not provided, the system will attempt to auto-detect all available sub-tools during initial connection to the MCP client and default to allowing all detected sub-tools. If you wish to restrict AI assistants to using only specific sub-tools, please list them explicitly here.

After clicking, the system will automatically fetch the tool content connected to that server and display it in the list:

MCP defined tool names

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 Reminders

Connection Testing

• After creating the tool, it's recommended to test if the MCP connection is working properly

• Verify tool functionality in the test environment

Permission Management

• Carefully select allowed tools to avoid authorizing unnecessary functions

• Regularly check tool usage status

Troubleshooting

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

• Confirm environment variables and parameter formats meet requirements