Happy slice of mcp pizzaMCP.pizza

deep-research-mcp

MCP.Pizza Chef: teelaitila

Point it at a topic and it writes its own follow-up searches, reads the pages it finds, scores how trustworthy each source looks, and returns a written report with findings and references. You set how wide and how deep it goes, and you can tell it to avoid listicles or affiliate reviews. Setup suits the technically comfortable: clone the code, build it, and supply your own paid keys for a search service and a model provider. Last updated January 2026.

Web/Research
Writing

Use This MCP server To

Research a topic and get a written report with sources Compare several products before I buy one Gather background on a company before a meeting Get a summary that flags which sources look shaky Skip listicles and affiliate reviews in my research

README

DISCLAIMER

This repo is an experiment on agent coding. 95% of the code is written by LLM's

Open Deep Research MCP Server

An AI-powered research assistant that performs deep, iterative research on any topic. It combines search engines, web scraping, and AI to explore topics in depth and generate comprehensive reports. Available as a Model Context Protocol (MCP) tool or standalone CLI. Look at exampleout.md to see what a report might look like.

Quick Start

  1. Clone and install:
git clone https://github.com/Ozamatash/deep-research
cd deep-research
npm install
  1. Set up environment in .env.local:
# Copy the example environment file
cp .env.example .env.local
  1. Build:
# Build the server
npm run build
  1. Run the cli version:
npm run start
  1. Test MCP Server with Claude Desktop:
    Follow the guide thats at the bottom of server quickstart to add the server to Claude Desktop:
    https://modelcontextprotocol.io/quickstart/server

For remote servers: Streamable HTTP

npm run start:http

Server runs on http://localhost:3000/mcp without session management.

Features

  • Performs deep, iterative research by generating targeted search queries
  • Controls research scope with depth (how deep) and breadth (how wide) parameters
  • Evaluates source reliability with detailed scoring (0-1) and reasoning
  • Prioritizes high-reliability sources (≥0.7) and verifies less reliable information
  • Generates follow-up questions to better understand research needs
  • Produces detailed markdown reports with findings, sources, and reliability assessments
  • Available as a Model Context Protocol (MCP) tool for AI agents
  • For now MCP version doesn't ask follow up questions
  • Natural-language source preferences (avoid listicles, forums, affiliate reviews, specific domains)

Model Selection (OpenAI, Anthropic, Google, xAI)

deep-research-mcp FAQ

Is this still maintained?
Loosely. The last code change was January 2026 and it is not archived. The author also describes the repository as an experiment in which a model wrote most of the code.
Do I need to pay for anything?
Yes. You supply your own keys for a web search service and for a model provider such as OpenAI, Anthropic, Google, or xAI.
How hard is setup?
You clone the code, install it, build it, and edit a settings file, so some comfort with a terminal is required.
Can I use this to research a topic and get a written report?
Yes, that is exactly what it does. The result comes back as a formatted report with its sources listed.
Will it ask me follow-up questions?
The command-line version does. The version that plugs into your chat app currently does not.
Which apps does it work in?
Claude Desktop is the documented setup. It can also run as a local web service that other apps can reach.
Can I avoid paying a search company?
Yes. You can run the free self-hosted crawler the project points to instead.