pminervini/deep-research-mcp

Overview

An MCP server that provides a unified interface for multiple deep research AI tools including OpenAI's Deep Research APIs, Gemini Deep Research Agent, Allen AI's DR-Tulu, and Hugging Face's Open Deep Research. It is written in Python and implements the Model Context Protocol to enable integration with compatible applications. Developers can route requests to different research models through a single MCP endpoint.

Best for

Best for
Developers building MCP-based tools that need flexible access to multiple deep research AI models

Use cases

• Connecting MCP-compatible applications to OpenAI Deep Research APIs
• Switching between Gemini Deep Research Agent and open-source alternatives via one server
• Building research automation tools that aggregate results from multiple deep research models

How to use

Install

pip install -r requirements.txt

Tools exposed

• uv
• pip

Tested with

Claude Code, Claude Desktop

Example client config

[research]\nprovider = "openai"\napi_style = "responses"\nmodel = "o4-mini-deep-research-2025-06-26"\napi_key = "your-api-key"\nbase_url = "https://api.openai.com/v1"\n[clarification]\nenable = true\ntriage_model = "gpt-5-mini"\nclarifier_model = "gpt-5-mini"\ninstruction_builder_model = "gpt-5-mini"

Notes

An MCP server that provides a unified interface for multiple deep research AI tools including OpenAI’s Deep Research APIs, Gemini Deep Research Agent, Allen AI’s DR-Tulu, and Hugging Face’s Open Deep Research. It is written in Python and implements the Model Context Protocol to enable integration with compatible applications. Developers can route requests to different research models through a single MCP endpoint.

84 stars on GitHub. Last updated 2026-05-13. Licensed MIT.

Use cases

• Connecting MCP-compatible applications to OpenAI Deep Research APIs
• Switching between Gemini Deep Research Agent and open-source alternatives via one server
• Building research automation tools that aggregate results from multiple deep research models

Pros

• Unifies access to proprietary and open-source deep research APIs
• Open-source Python implementation enables customization and self-hosting
• Supports multiple providers without changing client code

Cons

• Relatively low community adoption with 84 GitHub stars
• Requires MCP-compatible clients or adapters to use
• Each provider requires its own API keys and may have separate rate limits

Indexed from awesome-mcp-servers-punkpeye and enriched against its public facts.