TRTC Migration Assistant

TRTC Migration Assistant 是 TRTC云助手 团队为 腾讯云TRTC 打造的快速迁移MCP工具，可以在AI的帮助下快速将集成其他RTC厂商的业务代码迁移到腾讯云TRTC。

1. 功能

在支持MCP的AI工具中，选择或复制集成其他RTC厂商的业务代码，然后使用如下提示词提问

"xxxxxx" 这是xx友商的代码，帮我迁移到TRTC

该工具会解析源代码中对其他RTC厂商的接口调用，并提供迁移到TRTC对应的接口，文档和迁移指引，AI会根据这些信息进行准确的迁移，写出迁移后的代码。

2. 准备事项

2.1 AI工具

准备一个支持MCP的AI工具，例如CodeBuddy, Cursor, VSCode等

2.2 访问密钥

该工具通过腾讯云的云API提供服务，云API通过签名机制保证数据传输的安全性，该方式需要您在腾讯云申请安全凭证，即SecretID和SecretKey。

同时为了最小化该安全凭证的权限，推荐您在腾讯云控制台新建一个子用户，并只对该用户授予QcloudTrtccopilotFullAccess权限(该权限是用于访问TRTC云助手)，并在密钥管理处获取上述子用户的SecretID和SecretKey用于访问该服务。

3. 安装使用

安装该MCP有2种方式，分别是本地配置和使用托管的在线服务

3.1 本地配置

该方式需要您本地电脑有安装Python的包管理工具uv

然后在AI工具的MCP页面做如下配置，在TENCENTCLOUD_SECRET_ID和TENCENTCLOUD_SECRET_KEY处填入第2步获取的安全凭证(对应SecretID和SecretKey)。

{
  "mcpServers": {
    "trtc-migration-assistant": {
      "command": "uvx",
      "args": [
        "trtc-migration-assistant"
      ],
      "env": {
        "TENCENTCLOUD_SECRET_ID": "",
        "TENCENTCLOUD_SECRET_KEY": ""
      }
    }
  }
} 

3.2 使用托管服务

如果您不方便在自己的电脑上安装uv，本服务已在腾讯云开发者平台托管，您可以配置在线的连接

该方式需要您在腾讯云开发者平台找到trtc-migration-assistant，在页面右侧连接服务的部分输入前面获取到的SecretID和SecretKey，点击连接Server获取对应的MCP Server配置，例如：

{
  "mcpServers": {
    "trtc-migration-assistant": {
      "type": "sse",
      "url": "xxx"
    }
  }
}

注意上面的JSON仅为示例，需要您在腾讯云开发者平台获取自己的配置，然后将该配置设置到AI工具中即可

TRTC Migration Assistant

TRTC Migration Assistant is a rapid migration MCP tool built by the TRTC Copilot team for Tencent Cloud TRTC. With AI assistance, it helps quickly migrate source code integrated with other RTC vendors to Tencent Cloud TRTC.

1. Features

In an AI tool that supports MCP, select or copy source code integrated with other RTC vendors, then ask using the following prompt:

"xxxxxx" This is code from xx competitor/vendor. Please help me migrate it to TRTC.

The tool will parse interface calls to other RTC vendors in the source code and provide the corresponding TRTC interfaces, documentation, and migration guidance. Based on this information, the AI will accurately perform the migration and produce the migrated code.

2. Prerequisites

2.1 AI Tool

Prepare an AI tool that supports MCP, such as CodeBuddy, Cursor, or VSCode.

2.2 Access Keys

This tool provides services via Tencent Cloud’s Cloud API, which uses a signature mechanism to ensure data transmission security. You need to apply for security credentials on Tencent Cloud, namely SecretID and SecretKey.

To minimize the permissions of these credentials, it is recommended to create a sub-user in the Tencent Cloud Console, grant only the QcloudTrtccopilotFullAccess permission (used to access TRTC Copilot), and then retrieve the sub-user’s SecretID and SecretKey in Key Management for accessing this service.

3. Installation and Usage

There are two ways to set up this MCP: local configuration and using the hosted online service.

3.1 Local Configuration

This approach requires the Python package manager uv to be installed on your local machine.

Then, configure the MCP page in your AI tool as follows, and fill in the TENCENTCLOUD_SECRET_ID and TENCENTCLOUD_SECRET_KEY with the security credentials obtained in Step 2.

{
  "mcpServers": {
    "trtc-migration-assistant": {
      "command": "uvx",
      "args": [
        "trtc-migration-assistant"
      ],
      "env": {
        "TENCENTCLOUD_SECRET_ID": "",
        "TENCENTCLOUD_SECRET_KEY": ""
      }
    }
  }
}

3.2 Using the Hosted Service

If installing uv on your own computer is inconvenient, the service is hosted on the Tencent Cloud Developer Platform, and you can configure an online connection.

Go to the Tencent Cloud Developer Platform, find trtc-migration-assistant, enter the previously obtained SecretID and SecretKey in the “Connect Service” section on the right, and click “Connect Server” to get the corresponding MCP Server configuration, for example:

{
  "mcpServers": {
    "trtc-migration-assistant": {
      "type": "sse",
      "url": "xxx"
    }
  }
}

Note: The JSON above is only an example. You need to obtain your own configuration from the Tencent Cloud Developer Platform and then set that configuration in your AI tool.