MCP Servers

模型上下文协议服务器、框架、SDK 和模板的综合目录。

X
Xuezhe MCP Demo
作者 @hh696-wq

xuezhe-mcp-demo

创建于 3/6/2026
更新于 about 11 hours ago
Repository documentation and setup instructions

Xuezhe MCP Demo

Academic Intelligence Tools for AI Agents (MCP)

A simple demo client for the Xuezhe MCP Server, providing academic intelligence capabilities such as:

  • Scholar Search
  • Expert Discovery
  • Institution Ranking
  • Talent Mobility
  • Research Intelligence

This project demonstrates how AI agents and applications can interact with an MCP server to access structured academic data.


MCP Endpoint

https://mcp.xuezhe.pro/mcp

Authentication header:

X-API-Key: your_api_key


Features

Scholar Intelligence

  • scholar_search
  • scholar_detail

Search global researchers by:

  • name
  • institution
  • country

Expert Discovery

  • talent_discovery
  • talent_future_star
  • talent_doctor

Find experts and rising researchers by research topic.


Institution Intelligence

  • institution_ranking
  • institution_growth
  • institution_domain_structure

Analyze global universities, research institutes, and corporate research labs.


Talent Intelligence

  • talent_mobility

Track researcher movement between institutions and countries.


Research Intelligence

  • expert_map
  • corporate_intel
  • national_tech_structure

Understand global research ecosystems.


Installation

Clone the repository:

git clone https://github.com/hh696-wq/xuezhe-mcp-demo.git cd xuezhe-mcp-demo

Install dependencies:

pip install -r requirements.txt


Quick Start

Run the demo:

python mcp_demo.py --api-key YOUR_API_KEY --run-examples


Example Usage

List available tools:

python mcp_demo.py --api-key YOUR_API_KEY --list-tools

Call a tool:

python mcp_demo.py --api-key YOUR_API_KEY --tool institution_ranking --args '{"mode":"global","limit":5}'


Example Tools

Current MCP tools include:

scholar_search scholar_detail talent_discovery talent_mobility talent_doctor talent_future_star expert_map corporate_intel institution_ranking institution_growth institution_domain_structure national_tech_structure


MCP Session Flow

The demo follows the standard MCP workflow:

  1. initialize
  2. notifications/initialized
  3. tools/list
  4. tools/call

Example Use Cases

This MCP server enables AI agents to:

  • discover global research experts
  • analyze institutional research capabilities
  • identify rising scientific talent
  • understand global research trends

License

MIT

快速设置
此服务器的安装指南

安装包 (如果需要)

uvx xuezhe-mcp-demo

Cursor 配置 (mcp.json)

{ "mcpServers": { "hh696-wq-xuezhe-mcp-demo": { "command": "uvx", "args": [ "xuezhe-mcp-demo" ] } } }