Codealive

🚀 CodeAlive MCP Server Configuration Guide

CodeAlive MCP Server is a middleware tool designed to integrate with various AI clients. This document will guide you on how to configure the MCP Server to support different AI clients, including Claude Desktop, Cursor, and the Continue/Copilot plugins in VS Code.

🚀 Quick Start

Activate the Virtual Environment

Before running the MCP Server, ensure that you have activated your Python virtual environment:

source venv/bin/activate  # For macOS/Linux users
# Or for Windows users, use PowerShell to navigate to the project directory and run:
# .\venv\Scripts\activate

Start the MCP Server

In the virtual environment, set the CODEALIVE_API_KEY environment variable (replace it with your actual API key), and then start the MCP Server:

export CODEALIVE_API_KEY="your_api_key_here"
python src/codealive_mcp_server.py --debug

✨ Client Configuration

Claude Desktop Configuration

1. Open the settings interface of Claude Desktop.
2. Navigate to the "MCP" or "Extensions" section and add a new MCP Server:
   • Name: codealive
   • Type: CodeAliveMcpProvider
3. Fill in the following configuration parameters:

   {
     "command": "path/to/venv/bin/python",
     "args": [
       "src/codealive_mcp_server.py",
       "--debug"
     ],
     "env": {
       "CODEALIVE_API_KEY": "your_api_key_here"
     }
   }
   

Cursor Configuration

1. Open the settings of Cursor.
2. Navigate to the "MCP" section and add a new server configuration:
   • Name: codealive
   • Path: Point to your Python executable file (e.g., /path/to/venv/bin/python).

Continue/Copilot Plugins in VS Code

1. Open the settings of VS Code.
2. Search for and enter the configuration interface of "Continue" or "GitHub Copilot".
3. Add the MCP Server configuration:

   {
     "mcp.servers": [
       {
         "name": "codealive",
         "type": "CodeAliveMcpProvider",
         "command": "path/to/venv/bin/python",
         "args": ["src/codealive_mcp_server.py", "--debug"],
         "env": {
           "CODEALIVE_API_KEY": "your_api_key_here"
         }
       }
     ]
   }
   

🔧 Troubleshooting

Enable Debug Logging

Add the --debug flag to the client configuration to enable detailed logging:

{
  "args": [
    "src/codealive_mcp_server.py",
    "--debug"
  ]
}

Check MCP Server Output

Try to run the MCP Server command directly to ensure there are no error messages. Pay attention to the following:

• Ensure that the CODEALIVE_API_KEY is correct.
• Check for connection issues or other exceptions.

Check Client Logs

Refer to the relevant log files according to the client you are using:

• Claude Desktop:
  • macOS: /Users/your_username/Library/Logs/Claude/mcp.log
  • Windows: C:\Users\your_username\AppData\Local\Claude\logs\mcp.log
• Cursor: Check the output panel or console of VS Code.
• VS Code: Check the logs in the "Developer Tools".

Verify Environment Variables

Ensure that the CODEALIVE_API_KEY environment variable is correctly set in the client configuration without spelling errors or other formatting issues.

📄 License

The CodeAlive MCP Server is licensed under the MIT License. Please refer to the project repository for the full license terms and conditions.

If you encounter any issues during the configuration process, please refer to the official documentation or contact the support team (support@codealive.dev).