MCP Server Research Tool

@atimmeny27

Basic information

About MCP Server Research Tool

Uses an MCP server, API key, and claude to search thousands of sources and compile any topic into neatly formatted notes in markdown.

#mathgpt #math-solver #math-assistant

MCP Server Research Tool config

Add this server to your MCP-compatible client using the configuration below.

{
  "mcpServers": {
    "MCP-Server-Research-Tool": {
      "command": "python3",
      "args": [
        "-m",
        "ensurepip",
        "--upgrade"
      ]
    }
  }
}

MCP Server Research Tool tools

Tool list not available yet

This listing does not currently have a tool list to display. This does not mean the server has no tools; check its documentation for supported capabilities.

We auto-extract tools from the README. The maintainer can list them under a ## Tools heading to populate this section.

Overview

What is MCP-Server-Research-Tool?

MCP-Server-Research-Tool is a long-form research assistant powered by LLMs. It gathers information from primary sources, podcasts, academic PDFs, textbooks, YouTube, and more, then outputs structured markdown files with a detailed topic summary and source list.

How to use MCP-Server-Research-Tool?

Clone the repository, run the installation commands, set your OpenRouter API key (exported in .zshrc or .env), then launch with ./start.sh or manually with python MCP_assistant.py "Topic" and MCP_DURATION=long python send_to_claude.py. Enter a specific research topic and wait up to two minutes for the results.

Key features of MCP-Server-Research-Tool

Gathers information from primary sources, podcasts, and academic PDFs
Outputs structured markdown with summary and source list
Powered by LLMs via OpenRouter API (free key required)
Configure research prompt by editing send_to_claude.py lines 18-220
Supports manual launch or automated start script
Designed for macOS, can be added as a Desktop tool via Automator

Use cases of MCP-Server-Research-Tool

Conducting deep, realistic research on a specific historical or technological topic
Compiling structured notes from multiple media types (audio, video, text)
Creating markdown notes for use in editors like Obsidian
Learning about a subject by simulating LLM-driven multi-source investigation

Frequently asked questions

What API key is required?

You need a free API key from OpenRouter (or optionally from OpenAI/Anthropic). The key is set as environment variable `OPENROUTER_API_KEY` in `~/.zshrc` or a local `.env` file.

How long does research take?

The tool waits up to 2 minutes for the LLM to research the topic. You will see a message like “🔐 Using key: sk-or-v1-2 …” when it is actively researching.

What is the output?

The output is a markdown (.md) file containing a detailed summary and a list of sources used. It is best opened in a markdown-friendly editor such as Obsidian.

Can I edit the research prompt?

Yes. The instructions sent to Claude are in `send_to_claude.py` around lines 18-220 as a long f-string. You can modify that to change the research behavior.

Are there any known limitations?

Wikipedia’s API may be overloaded, causing temporary errors. Spelling errors or niche topics may confuse the LLM. For best results, use a specific, detailed research topic.

Comments

More Data & Analytics MCP servers

OAIA Arena

Give any AI trading agent a public, un-cherry-pickable track record. Commit a call before it resolves, get a hash receipt, get scored against 2,048 coin-flipping monkeys on the same window.

CompanyData

Company search and enrichment for AI agents: 200+ countries of company data from official business registers. Search companies, get full firmographic records and look up exact city names. Hosted MCP, no install.

TikTok MCP Server

HasData

Hosted TikTok MCP server for AI agents: read public profiles, account videos, comments and replies, or search videos and creators. Structured JSON for creator research and content analysis.

EximAgent

AI sales team for export-import businesses: specialist agents for outbound prospecting, trade data analysis and compliance (HS codes, tariffs, landed cost, OFAC sanctions).

AdLoop Cloud

kLOsk

Hosted AdLoop: Google Ads, Reddit Ads, GA4 and Tag Manager in claude.ai, ChatGPT and more, with a preview before every change.