Deep Research MCP server

This MCP server provides a multi-agent research framework that coordinates various AI agents to perform complex research tasks using the Message Control Protocol. When connected to an MCP-compatible client, you can use the deep_research tool to automate research workflows.

Installation

Prerequisites

• Python 3.8 or higher
• An OpenAI API key
• An MCP-compatible client (like Claude Desktop)

Installation Steps

1. Clone the repository

   git clone https://github.com/yourusername/deep-research-mcp.git
   cd deep-research-mcp
   

2. Set up environment variables

   cp .env.example .env
   

   Edit the .env file to add your API keys and other required variables:

   OPENAI_API_KEY=sk-123-yourkey
   # Add any additional environment variables here
   

3. Install required dependencies

   pip install -r multi_agents/requirements.txt
   

4. Configure your MCP client

   Edit your claude_desktop_config.json file to include:

   {
       "mcpServers": {
           "deep-research-mcp": {
               "command": "path/to/your/python/interpreter",
               "args": [
                   "/path/to/this/project/deep-research-mcp/mcp_server.py"
               ]
           }
       }
   }
   

   Be sure to replace the paths with the actual paths on your system.

Using the MCP Server

Starting the Server

Run the MCP server with:

python mcp_server.py

This starts a FastMCP server locally, exposing the deep_research tool to any connected MCP-compatible client.

Connecting to the Server

With the server running, open your MCP-compatible client (such as Claude Desktop). The deep_research tool should now be available through the client's tool interface.

Using the Deep Research Tool

The tool coordinates multiple specialized agents:

• ResearchAgent: Gathers information from specified sources
• EditorAgent: Organizes and structures the research
• WriterAgent: Creates content based on the research
• And others depending on your task

Each agent plays a specific role in the research workflow, allowing for complex tasks to be broken down and executed efficiently.

Example Usage

When using the tool through an MCP-compatible client, you'll typically:

1. Specify a research task or question
2. Set parameters for the research scope and depth
3. The system will coordinate the agents to:
   • Gather relevant information
   • Organize findings
   • Generate a structured output

The final result can be returned directly to your client or saved to files depending on the configuration.

Troubleshooting

• Connection issues: Ensure the server is running before attempting to connect
• Authentication errors: Verify your API keys in the .env file
• Missing dependencies: Check that all requirements are installed correctly

For more complex errors, check the server logs which display in the terminal where you ran the server.

How to add this MCP server to Cursor

There are two ways to add an MCP server to Cursor. The most common way is to add the server globally in the ~/.cursor/mcp.json file so that it is available in all of your projects.

If you only need the server in a single project, you can add it to the project instead by creating or adding it to the .cursor/mcp.json file.

Adding an MCP server to Cursor globally

To add a global MCP server go to Cursor Settings > MCP and click "Add new global MCP server".

When you click that button the ~/.cursor/mcp.json file will be opened and you can add your server like this:

{
    "mcpServers": {
        "cursor-rules-mcp": {
            "command": "npx",
            "args": [
                "-y",
                "cursor-rules-mcp"
            ]
        }
    }
}

Adding an MCP server to a project

To add an MCP server to a project you can create a new .cursor/mcp.json file or add it to the existing one. This will look exactly the same as the global MCP server example above.

How to use the MCP server

Once the server is installed, you might need to head back to Settings > MCP and click the refresh button.

The Cursor agent will then be able to see the available tools the added MCP server has available and will call them when it needs to.

You can also explictly ask the agent to use the tool by mentioning the tool name and describing what the function does.