Unichat Ts MCP Server

🚀 Unichat MCP Server (TypeScript Version)

This project offers a TypeScript implementation of the Unichat MCP server. There's also a Python version available.

This server allows you to send requests to OpenAI, MistralAI, Anthropic, xAI, Google AI, or DeepSeek using tools or preset prompts via the MCP protocol. It requires the API keys of the respective providers. It supports both STDIO and SSE transmission mechanisms, depending on the parameters.

✨ Features

Tools

The server implements one tool:

• unichat: Sends requests to Unichat.
  • Accepts "messages" as a required string parameter.
  • Returns the response.

Preset Prompts

• code_review
  • Reviews code to identify best practices, potential issues, and improvement suggestions.
  • Parameters:
    • code (string, required): The code to be reviewed.
• document_code
  • Generates documentation for the code, including comments and docstrings.
  • Parameters:
    • code (string, required): The code to be documented.
• explain_code
  • Explains in detail how the code works.
  • Parameters:
    • code (string, required): The code to be explained.
• code_rework
  • Modifies the provided code according to the requirements.
  • Parameters:
    • changes (string, optional): The changes to be applied.
    • code (string, required): The code to be modified.

📦 Installation

Installation via Smithery

To automatically install the Unichat MCP server for Claude Desktop via Smithery, use the following command:

npx -y @smithery/cli install unichat-ts-mcp-server --client claude

Manual Installation

To use it with Claude Desktop, add the server configuration:

• On MacOS: ~/Library/Application Support/Claude/claude_desktop_config.json
• On Windows: %APPDATA%/Claude/claude_desktop_config.json

Run locally:

{
  "mcpServers": {
    "unichat-ts-mcp-server": {
      "command": "node",
      "args": ["index.js"],
      "path": "{Server Path}",
      "env": {
        "NODE_ENV": "development"
      }
    }
  }
}

Install via the following command:

npm install unichat-mcp-server --save

💻 Usage Examples

Preset Prompts

Set the preset prompts:

const config = {
  prompt: {
    code_review: "As a code review expert, analyze the following code in detail and provide improvement suggestions: {code}",
    document_code: "Generate comments and documentation for the following code: {code}",
    explain_code: "Explain in detail how the following code works: {code}",
    code_rework: "Modify the following code according to the requirements: {changes}"
  }
};

Tools

Call the tool:

const server = new MCP.Server();
server.use(new MCP.Tool('unichat'));
await server.start();

🔧 Technical Details

Development

Install dependencies:

npm install

Build the server:

npm run build

Develop with automatic rebuild:

npm run watch

Configuration

Add the following to the .env file:

MCP_SERVER_MODE=development
PORT=3000

📄 License

No license information provided in the original document.

⚠️ Important Note

• Keep the code concise and maintainable.
• Ensure security and stability.
• Regularly update dependencies.