MCP-GTP-researcher

A Model Context Protocol server that gives AI assistants the ability to conduct systematic, multi-source research on any topic. This is not a search tool — it's a research infrastructure.

GPT Researcher is an autonomous research agent that doesn't just find information — it investigates. It spawns multiple agents to explore different angles of a question, validates sources across the web, and synthesizes findings into comprehensive reports with proper citations.

• Source: github.com/assafelovic/gpt-researcher
• Architecture: Multi-agent system with planner and crawler agents
• Output: Detailed reports with sourced claims, not vibes

The official project describes itself as "an autonomous agent for comprehensive online research."

It exposes GPT Researcher's capabilities through the Model Context Protocol. Your AI assistant becomes a research lead that can delegate deep investigation tasks, receive structured reports, and build on previous research.

| Operation | Description | Latency | |-----------|-------------|---------| | deep_research | Full comprehensive research report | 3-5 min | | quick_research | Rapid assessment with sources | 30-60 sec |

1. GPT Researcher running somewhere

For installation, see the official deployment guide.

1. Python environment

git clone https://github.com/assafelovic/mcp-gpt-researcher.git
cd mcp-gpt-researcher
python -m venv .venv
source .venv/bin/activate
pip install -r requirements.txt

1. Configuration

cp .env.example .env

Set GPT_RESEARCHER_BASE_URL to your GPT Researcher instance. The default port is 10305.

| Variable | Default | Description | |----------|---------