TCSAS-Devtools MCP Server

An MCP (Model Context Protocol) server designed to connect AI assistants with TCSAS-Devtools for a seamless miniprogram development experience.

This server wraps the basic functionalities of TCSAS-Devtools into a set of tools that can be called by AI assistants (like CodeBuddy, Cursor, Claude), allowing you to open, build, preview, and upload miniprograms using natural language.

✨ Features

• Launch IDE: Directly open the TCSAS-Devtools application or a specific miniprogram project.
• Check Installation: Verify if TCSAS-Devtools is installed on the local machine.
• Preview Miniprogram: Generate a QR code for previewing the miniprogram on a mobile device.
• Upload Miniprogram: Upload a new version of the miniprogram with specified version information.
• Get Runtime Log: Retrieve the latest runtime log and screenshot from the MiniProgram IDE for debugging.

🚀 Prerequisites

Before you begin, ensure you have the following software installed:

• Node.js (v20 or higher recommended)
• npm
• TCSAS-Devtools

📦 Installation and Setup

1. For first-time use, please download the installation package for TCSAS-Devtools from the console in advance, the version should >= 2.2.15.
2. After installation, please open it to complete the login process and enable the service port.

Settings - Security - Enable

Ensure the above steps are completed. Then you can configure the MCP server in your Code Agent and start using it.

🤖 How to Use in Cursor

You can integrate this server with any AI assistant that supports custom MCP servers. Here are the steps to configure it in Cursor:

1. Open Cursor Settings: In Cursor, find the settings for AI integration. This is usually located in Settings > Tools & MCP or a similar configuration area for managing AI providers.

2. Add Custom MCP Service: Look for an option like "Add Model Context Provider" or "MCP Server" and click "Add".

3. Enter Command: For example:

   {
     "tcsas-devtools-mcp-server": {
       "command": "npx",
       "args": [
         "-y",
         "tcsas-devtools-mcp-server@latest"
       ]
     }
   }

   Illustration: (A screenshot of the Cursor settings panel could be like this, showing the user entering the command in the mcp.json)

   

4. Connect and Start Using: After adding, Cursor will connect to your local server. Now you can interact with it in the chat. For example, you can type:

   "Launch TCSAS-Devtools and open my project at /Users/me/projects/my-miniprogram."

   "Generate a preview QR code for the project I'm currently developing."

🛠️ Tool API Reference

Below is a detailed description of all the tools provided by this MCP server:

launchIde

Launches the TCSAS-Devtools IDE.

• Input Parameters:
  • path (optional, string): The absolute path to the miniprogram project to open.
  • ideInstallPath (optional, string): The absolute path to the IDE installation. If omitted, the IDE will be launched with the default installation path.
• Output:
  • openApp (boolean): Whether the IDE was launched successfully.
  • openProject (boolean): Whether the project was opened successfully.
  • msg (string): Status message or error details.

checkIdeInstalled

Checks if the TCSAS-Devtools IDE is installed.

• Input Parameters: None
• Output:
  • isInstall (boolean): Returns true if the IDE is installed, otherwise false.

previewMiniprogram

Generates a preview QR code for a miniprogram project. Note: This process can take 60-80 seconds to complete.

• Input Parameters:
  • path (string): The absolute path to the miniprogram project.
• Output:
  • Returns a QR code image (PNG) that can be scanned to preview the miniprogram.

uploadMiniprogram

Uploads a new version of the miniprogram.

• Input Parameters:
  • path (string): The absolute path to the project.
  • version (string): The new version number (e.g., "1.0.1").
  • describeMessage (string): A short description of the changes.
• Output:
  • updateDetail (string): Upload response details or error message.

getMiniProgramRuntimeLog

Get the latest runtime log and screenshot from MiniProgram IDE. Use this after writing or modifying code to verify execution results and UI rendering.

• Input Parameters:
  • path (string): The absolute path of the miniprogram project.
• Output:
  • result (string): The runtime log from MiniProgram (console.log, warnings, errors).
  • timestamp (string): When the log was generated.
  • screenshot (image/png): A screenshot of the current MiniProgram view.

🤝 Contributing

Contributions are welcome! Please feel free to submit a Pull Request.

📜 License

This project is open-sourced under the MIT License.