Twosplit

🚀 Two-Split MCP Server

A Multi-Claude Processor (MCP) server that leverages multiple Claude instances to provide enhanced responses. It sends the same prompt to two different Claude instances and uses a third instance to combine or select the best elements.

✨ Features

• Support for multiple Claude models:
  • claude-3-opus-latest
  • claude-3-5-sonnet-latest
  • claude-3-5-haiku-latest
  • claude-3-haiku-20240307
• Obtain single direct responses from each AI
• Display the original responses and source attributions
• Return an optimized final response

📦 Installation

1. Clone the repository.
2. Install dependencies:

npm install

3. Build the server:

npm run build

Configuration

The server requires an Anthropic API key to run. Set it as an environment variable:

export ANTHROPIC_API_KEY=your-api-key-here

💻 Usage Examples

Basic Usage

The server provides a tool named twosplit with the following parameters:

• prompt (required): The prompt to send to Claude
• model (required): The Claude model to use (must be one of the supported models in the list)

Example of using the tool with Claude:

<use_mcp_tool>
<server_name>twosplit</server_name>
<tool_name>twosplit</tool_name>
<arguments>
{
  "prompt": "Write a short story about a robot learning to paint",
  "model": "claude-3-5-sonnet-latest"
}
</arguments>
</use_mcp_tool>

The response will include:

1. The final optimized response
2. The original responses from the two AIs
3. Display of partial source attributions

🔧 Technical Details

How It Works

1. The server sends the same prompt to two different instances of the specified Claude model, requesting single direct responses.
2. A third instance analyzes the two responses and:
   • Selects the best response if one is clearly superior.
   • Creates a new response by combining the best elements of the two responses.
3. The final response, original responses, and source attributions are all included in the output.

Development

Run the server in development mode

npm run watch

Check the server's functionality

npm run inspector