home / mcp / ai collaboration mcp server

AI Collaboration MCP Server

This is a very basic MCP Server for VS Code to use with Copilot to enhance Copilot knowledge and ability.

Installation

Add the following to your MCP client configuration file.

Configuration

{
  "mcpServers": {
    "hurryupmitch-ai-collaboration-mcp-server": {
      "command": "node",
      "args": [
        "/path/to/ai-collaboration-mcp-server/build/index.js"
      ],
      "env": {
        "OLLAMA_MODEL": "llama3.2:latest",
        "GEMINI_API_KEY": "YOUR_KEY_HERE",
        "OPENAI_API_KEY": "YOUR_KEY_HERE",
        "OLLAMA_BASE_URL": "http://localhost:11434",
        "ANTHROPIC_API_KEY": "YOUR_KEY_HERE"
      }
    }
  }
}

The AI Collaboration MCP Server provides a focused set of tools to streamline AI-assisted development inside your editor. It maintains workspace-specific context, automatically injects project details, and lets you query multiple AI providers with centralized tooling. Use it to get expert AI guidance, compare perspectives, and run actions directly from your IDE.

How to use

You interact with the MCP server through your MCP-enabled editor client. Start by configuring a local or workspace-specific MCP server entry, then use the four core tools to manage your project context and AI interactions.

How to install

Prerequisites: Node.js 18+ and a editor with MCP support. You also need API keys for the AI providers you plan to use.

Additional sections

Configuration notes, security considerations, troubleshooting tips, and development commands are detailed below for a smooth setup and ongoing maintenance.

Security and troubleshooting

Keep API keys secure in your MCP configuration. Conversation history is stored locally on your machine. If you encounter startup or authentication issues, verify your configuration, review the MCP server logs, and ensure your editor extension is up to date.

Available tools

#set_workspace

Set the current workspace directory to enable project-specific conversation history and automatic context injection for that project.

#consult_ai

Ask for expert guidance from a specific AI provider with full project context.

#multi_ai_research

Obtain perspectives from multiple AI providers on a given topic to compare approaches.

#mandatory_execute

Force tool execution with explicit commands when needed.