Deep Research

Provides a multi-agent research framework using Python and MCP to orchestrate AI agents for research tasks.

python

10

GitHub Stars

python

Language

6 months ago

First Indexed

2 months ago

Catalog Refreshed

Documentation & install

Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.

Installation

Add the following to your MCP client configuration file.

Configuration

View docs

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

Deep Research MCP is an automated, multi-agent research framework that runs as a FastMCP server and exposes a dedicated tool named deep_research. It enables you to orchestrate AI agents for research tasks, draft outputs, and iterative collaboration through MCP clients or CLIs.

How to use

Use an MCP client or the CLI to invoke the deep_research tool exposed by the FastMCP server. You will interact with a single entry point that coordinates multiple agents to perform research tasks, generate drafts, and refine results within a structured memory of ongoing work.

How to install

Prerequisites and initial setup help you get started quickly.

Install Python and a compatible package manager on your machine.

Concrete steps you should follow to set up the server locally:

# 1) Clone the project
git clone https://github.com/yourusername/deep-research-mcp.git
cd deep-research-mcp

# 2) Create your environment file
cp .env.example .env
# Edit the new .env file to fill in your secrets and environment variables
# For example:
# OPENAI_API_KEY=sk-123-yourkey

# 3) Install dependencies
pip install -r multi_agents/requirements.txt

# 4) Prepare configuration for MCP (example entry shown in code block below)
# Edit claude_desktop_config.json to include the MCP server entry as shown in the snippet

Configuring the MCP server and running

Next, configure the MCP server entry so your client can discover and start the deep_research tool. The server uses a stdio-based setup pointing to the Python interpreter and the mcp_server entry script.

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

Run the MCP server

Start the FastMCP server locally so clients can connect to the deep_research tool.

python mcp_server.py

Available tools

deep_research

Exposed MCP tool that runs the deep research workflow by coordinating multiple agents to perform research tasks, draft outputs, and refine results.