home / mcp / open deep research mcp server

Open Deep Research MCP Server

Provides deep, iterative research with source reliability scoring and markdown reports for MCP-enabled AI agents.

Installation

Add the following to your MCP client configuration file.

Configuration

{
  "mcpServers": {
    "ozamatash-deep-research-mcp": {
      "url": "http://localhost:3000/mcp",
      "headers": {
        "XAI_API_KEY": "YOUR_XAI_API_KEY",
        "GOOGLE_API_KEY": "YOUR_GOOGLE_API_KEY",
        "OPENAI_API_KEY": "YOUR_OPENAI_API_KEY",
        "ANTHROPIC_API_KEY": "YOUR_ANTHROPIC_API_KEY"
      }
    }
  }
}

Open Deep Research MCP Server is an AI-powered research assistant that performs deep, iterative research on any topic. It combines search, data gathering, and AI processing to produce comprehensive reports, and can run as an MCP tool or a standalone CLI for flexible usage.

How to use

Use the MCP client to connect to the Open Deep Research MCP Server and start research sessions. The server can run locally or be accessed remotely via HTTP. It will generate targeted search queries, perform web scraping as needed, evaluate source reliability, and produce a structured Markdown report with findings, sources, and reliability assessments.

How to install

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

cp .env.example .env.local

npm run build

npm run start

npm run start:http

Additional notes

The server is designed to run on http://localhost:3000/mcp when using the HTTP method. A quick-start path includes cloning, installing dependencies, building, and starting the server, followed by optional testing with Claude Desktop by following the provided setup guidance.

Available tools

serp_query_generation

Generates targeted search queries to drive deep, iterative research.

search_and_collect

Performs web searches and collects sources for evaluation.

reliability_evaluation

Evaluates source reliability with detailed scores and reasoning.

report_generation

Produces detailed Markdown reports with findings, sources, and reliability assessments.

followup_question_generation

Creates follow-up questions to refine research directions.