Deep Research

Created By

Pink Pixel7 months ago

A Model Context Protocol (MCP) compliant server designed for comprehensive web research. It uses Tavily's Search and Crawl APIs to gather detailed information on a given topic, then structures this data in a format perfect for LLMs to create high-quality markdown documents.

# information-retrieval

# knowledge-base

Overview Content Tools Comments

Overview

Deep Research MCP Server ✨

Dream it, Pixel it — by Pink Pixel

🚀 Project Purpose

The Deep Research MCP Server is a Model Context Protocol (MCP) compliant server for comprehensive, up-to-date web research. It leverages Tavily's Search & Crawl APIs to gather, aggregate, and structure information for LLM-powered documentation generation.

🧩 Architecture Overview

MCP Server (Node.js, TypeScript)
Stdio Transport for agent/server communication
Tavily API Integration (Search + Crawl)
Configurable Documentation Prompt (default, ENV, or per-request)
Structured JSON Output for LLMs

Architecture Diagram (Text)

[LLM/Agent]
    │
    ▼
[Deep Research MCP Server]
    │   ├─> Tavily Search API
    │   └─> Tavily Crawl API
    ▼
[Aggregated JSON Output + Documentation Instructions]

✨ Main Features

Multi-Step Research: Combines AI-powered search with deep content crawling
Structured Output: JSON with query, search summary, findings, and doc instructions
Configurable Prompts: Override documentation style via ENV or per-request
Configurable Output Path: Specify where research documents and images should be saved
Granular Control: Fine-tune search/crawl with many parameters
MCP Compliant: Plug-and-play for agent ecosystems
Resource Optimized: Memory tracking, auto-garbage collection, and hardware acceleration support

🛠️ Key Dependencies

@modelcontextprotocol/sdk (v1.11.1) — MCP server framework
@tavily/core (v0.5.2) — Tavily Search & Crawl APIs
dotenv (v16.5.0) — Environment variable management

📁 File Structure

deep-research-mcp/
├── dist/                # Compiled JS output
├── src/
│   └── index.ts         # Main server logic
├── assets/              # Project assets (logo)
├── README.md            # Full documentation
├── OVERVIEW.md          # (You are here!)
├── example_config.json  # Example MCP config
├── package.json         # Project metadata & dependencies
├── tsconfig.json        # TypeScript config
├── CHANGELOG.md         # Version history and changes

⚡ Usage & Integration

Install & Run:
- npx @pinkpixel/deep-research-mcp (quickest)
- Or clone & npm install, then npm start
Configure: Set TAVILY_API_KEY in your environment (see README.md)
Integrate: Connect to your LLM/agent via MCP stdio
Customize: Override documentation prompt via ENV or tool argument
Output: Specify where research documents and images should be saved
Performance: Enable hardware acceleration with hardware_acceleration: true parameter

🔄 Recent Updates

Optimized Resource Usage: Reduced default crawl limits to prevent excessive memory consumption
MCP Protocol Compliance: Fixed response structure to properly follow MCP specifications
Improved Error Handling: Better error reporting and handling of timeouts
Performance Optimizations: Added optional hardware acceleration (WebGPU) support
Smarter Crawling: Added domain validation to focus crawling and prevent overly broad searches

See CHANGELOG.md for complete version history

📚 More Info

See README.md for full usage, parameters, and troubleshooting
Example config: example_config.json
License: MIT
Node.js: >=18.0.0 required

Made with ❤️ by Pink Pixel

Server Config

{
  "mcpServers": {
    "deep-research": {
      "command": "npx",
      "args": [
        "-y",
        "@pinkpixel/deep-research-mcp"
      ],
      "env": {
        "TAVILY_API_KEY": "tvly-YOUR_ACTUAL_API_KEY_HERE",
        "DOCUMENTATION_PROMPT": "Your custom, detailed instructions for the LLM on how to generate markdown documents from the research data...",
        "RESEARCH_OUTPUT_PATH": "/path/to/your/research/output/folder"
      }
    }
  }
}

Recommend Servers

TraeBuild with Free GPT-4.1 & Claude 3.7. Fully MCP-Ready.

Zhipu Web SearchZhipu Web Search MCP Server is a search engine specifically designed for large models. It integrates four search engines, allowing users to flexibly compare and switch between them. Building upon the web crawling and ranking capabilities of traditional search engines, it enhances intent recognition capabilities, returning results more suitable for large model processing (such as webpage titles, URLs, summaries, site names, site icons, etc.). This helps AI applications achieve "dynamic knowledge acquisition" and "precise scenario adaptation" capabilities.

TimeA Model Context Protocol server that provides time and timezone conversion capabilities. This server enables LLMs to get current time information and perform timezone conversions using IANA timezone names, with automatic system timezone detection.

WindsurfThe new purpose-built IDE to harness magic

CursorThe AI Code Editor

EdgeOne Pages MCPAn MCP service designed for deploying HTML content to EdgeOne Pages and obtaining an accessible public URL.

Baidu Map百度地图核心API现已全面兼容MCP协议，是国内首家兼容MCP协议的地图服务商。

Serper MCP ServerA Serper MCP Server

Visual Studio Code - Open Source ("Code - OSS")Visual Studio Code

DeepChatYour AI Partner on Desktop

Howtocook Mcp基于Anduin2017 / HowToCook （程序员在家做饭指南）的mcp server，帮你推荐菜谱、规划膳食，解决“今天吃什么“的世纪难题； Based on Anduin2017/HowToCook (Programmer's Guide to Cooking at Home), MCP Server helps you recommend recipes, plan meals, and solve the century old problem of "what to eat today"