一个MCP（模型上下文协议）服务器，支持Claude及其他兼容MCP的AI助手的全面研究能力。该服务器集成了网络和学术搜索功能，使AI模型能够访问来自多个来源的最新信息，跟踪相关链接，并提供结构良好的研究成果。

概述

Claude Deep Research 是一款强大的研究工具，通过提供以下功能来扩展大型语言模型的能力：

1. 通过DuckDuckGo实现的网页搜索集成
2. 通过语义学者获取学术研究
3. 网页内容提取
4. 结构化格式的全面分析
5. 数据表示的可视化指导

该服务器遵循MCP的设计原则，实现与Claude及其他AI助手的无缝集成。

特色

• 统一研究工具：网络和学术信息的单一界面
• 多源整合：将来自多个来源的信息整合成连贯的研究
• 内容提取：从网页中提取相关信息
• 学术资料发现：查找与你主题相关的学术文章

Claude Deep Research

An MCP (Model Context Protocol) server that enables comprehensive research capabilities for Claude and other MCP-compatible AI assistants. This server integrates web and academic search functionality, allowing AI models to access current information from multiple sources, follow relevant links, and provide well-structured research results.

Overview

Claude Deep Research is a powerful research tool that extends the capabilities of LLMs by providing:

1. Web search integration through DuckDuckGo
2. Academic research access through Semantic Scholar
3. Content extraction from web pages
4. Comprehensive analysis with structured formatting
5. Visualization guidance for data representation

The server follows MCP design principles to provide a seamless integration with Claude and other AI assistants.

Features

• Unified Research Tool: Single interface for web and academic information
• Multi-Source Integration: Combines information from various sources into cohesive research
• Content Extraction: Pulls relevant information from web pages
• Academic Source Discovery: Finds scholarly articles related to your topic
• Smart Formatting: Properly formats research with citations
• Visual Framework: Provides guidance for creating effective data visualizations
• Structured Analysis: Organizes research using academic methodologies

Installation

Prerequisites

• Python 3.8 or higher
• pip or uv package manager

Quick Install

# Using pip
pip install mcp httpx beautifulsoup4

# Clone the repository
git clone https://github.com/yourusername/claude-deep-research.git

Configuration

The server works out of the box with default settings, but you can modify the following parameters in deep_research.py for customization:

# Configuration
USER_AGENT = "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/91.0.4472.124 Safari/537.36"
MAX_CONTENT_SIZE = 8000  # Maximum characters in the final response
MAX_RESULTS = 3         # Maximum number of results to process

Usage

Running the Server

Modify your Claude desktop config and restart Claude. On a Mac this is at ~/Library/Application Support/Claude

  "search-scholar": {
      "command": "/python",
      "args": [
        "/deep_research.py"
      ]
    }

Using with Claude Desktop

Once installed, you can access the server in Claude Desktop:

1. Tool Access: Use the deep_research tool directly in conversation

Research Tool

The main deep_research tool accepts the following parameters:

• query (required): The research question or topic
• sources (optional): Which sources to use: “web”, “academic”, or “both” (default)
• num_results (optional): Number of sources to examine (default 2, max 3)

Example prompts: