Open Deep Research MCP server for AI agents

This MCP server functions as an AI-powered research assistant that performs deep, iterative research on any topic. It combines search engines, web scraping, and AI to explore topics thoroughly and generate comprehensive reports, available as a Model Context Protocol (MCP) tool or standalone CLI.

Installation

To install the Deep Research MCP server:

Clone the repository and install dependencies:

git clone https://github.com/Ozamatash/deep-research
cd deep-research
npm install

Set up the environment configuration:

# Copy the example environment file
cp .env.example .env.local

Build the server:
```
npm run build
```

Running the Server

CLI Version

Run the command-line interface version:

npm run start

MCP Server

To use it as an MCP server with Claude Desktop:

Start the server in HTTP mode:
```
npm run start:http
```
The server will run on http://localhost:3000/mcp without session management.
Add the server to Claude Desktop by following the guide at: https://modelcontextprotocol.io/quickstart/server

Features and Capabilities

Depth and Breadth Control: Customize research scope with depth (how deep) and breadth (how wide) parameters
Source Reliability: Evaluates sources with detailed scoring (0-1) and reasoning
Source Prioritization: Focuses on high-reliability sources (≥0.7) and verifies less reliable information
Detailed Reports: Produces markdown reports with findings, sources, and reliability assessments

Using Local Firecrawl (Free Option)

Instead of the Firecrawl API, you can run a local instance:

Set up local Firecrawl:

git clone https://github.com/Ozamatash/localfirecrawl
cd localfirecrawl
# Follow setup in localfirecrawl README

Update your .env.local file:

FIRECRAWL_BASE_URL="http://localhost:3002"

Optional Observability

You can add observability to track research flows, queries, and results using Langfuse:

Add to your .env.local file:

LANGFUSE_PUBLIC_KEY="your_langfuse_public_key"
LANGFUSE_SECRET_KEY="your_langfuse_secret_key"

The application works normally without observability if no Langfuse keys are provided.

How the Research Process Works

The deep research process follows these steps:

User provides a query and parameters for breadth and depth
System generates targeted search queries
Performs searches and evaluates source reliability
Processes results and assigns reliability scores
Recursively explores deeper based on depth parameter
Generates a comprehensive markdown report

This iterative approach allows for thorough exploration of complex topics with reliability assessment at each stage.

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.