qs MCP

QS World University Rankings MCP Server

This is a QS World University Ranking query server based on FastMCP, providing intelligent multi-dimensional query functions such as university search, ranking query, and national statistics.

Features

This MCP server provides the following nine tools:

1. search_university - University Search

Search for universities based on keywords to view their rankings and various indicators.

parameter

• keywordKeywords for university names (supports fuzzy matching, case insensitive), such as "MIT", "Peking", "Tsinghua"
• year: Year (optional, such as 2024, 2025, 2026), if not specified, all years will be displayed

return:

• Match university ranking information and various indicators
• Multi year ranking trend analysis

2. get_top_universities - Ranking Query

Search for the top N universities by year and country/region.

parameter

• yearYear (required, such as 2024, 2025, 2026)
• country: Country/region code or name (optional, such as CN, US, UK, China)
• top_nDisplay the top N names (default 10, maximum 100)

return:

• List of top N ranked universities
• Includes various scoring indicators

3. get_comcounty_stats - National Statistics

Count the number and distribution of universities by country/region.

parameter

• year: Year (required)
• top_nDisplay the top N countries (default 20)

return:

• Statistics on the number of universities in each country/region
• Number of ranked universities
• percentage share

4. get_country_stores - Comparison of National Average Scores

Calculate the average score by country/region.

parameter

• year: Year (required)
• top_nDisplay the top N countries (default 15)

return:

• Average score, highest score, and lowest score of each country
• Best ranked universities

5. get_rank_changes - Rank Changes

View the universities with the largest changes in rankings.

parameter

• yearYear (required, will be compared with the previous year)
• top_nDisplay the top N universities (default 20)
• directionChange direction, "rise" rising or "fall" falling

return:

• List of universities with the largest ranking changes
• Including a comparison of rankings between last year and this year

6. get_top100_dedistribution - Top 100 distribution

View the national distribution of the top 100 universities.

parameter

• year: Year (required)

return:

• Statistics on the National Distribution of Top 100 Universities
• Proportion of each country

7. get_Summary - Comprehensive Statistics

Obtain a comprehensive statistical summary for the specified year.

parameter

• year: Year (required)

return:

• Basic information (total number of universities, ranking, number of countries)
• Top 10 universities
• Top 5 countries
• score statistics
• Compared to last year

8. list\_ available years - View available years

List all available years in the data.

return:

• List of available years
• latest year

9. list_comountries - View country list

List all available countries/regions in the data.

parameter

• year: Year (optional)

return:

• List of available countries/regions

Data source

Using QS World University Ranking data (2024-2026), including the following indicators:

• Overall Score
• Academic Reputation
• Employer Reputation
• Faculty Student Ratio
• Citations per Faculty
• International Faculty (International Teachers)
• International Students
• International Research Network
• Employment outcomes
• Sustainability (Sustainable Development)

Installation and operation

local run

# 安装依赖
pip install -r requirements.txt

# 启动服务器
python mcp_server.py

Alibaba Cloud Function Computing Deployment

1. ensure qs_cleaned.csv The data file is located in the root directory of the project
2. Run the build script:

   chmod +x build.sh
   ./build.sh

1. Package and upload to Alibaba Cloud for function calculation
2. Set the startup command to ./run.sh

environment variable

• QS_CSV_PATH: Data file path (optional, automatically searched by default)
• PORT / MCP_PORTService port (default 9001)

API interface

The server runs in SSE (Server Send Events) mode, with the default listening address being:

• 本地：http://localhost:9001/sse
• Internethttp://0.0.0.0:9001/sse

Usage example

Search for universities

{
  "tool": "search_university",
  "arguments": {
    "keyword": "MIT",
    "year": 2026
  }
}

Search for the ranking of Chinese universities

{
  "tool": "get_top_universities",
  "arguments": {
    "year": 2026,
    "country": "CN",
    "top_n": 20
  }
}

View ranking changes

{
  "tool": "get_rank_changes",
  "arguments": {
    "year": 2026,
    "top_n": 10,
    "direction": "rise"
  }
}

project structure

qs_mcp/
├── mcp_server.py      # MCP 服务器主程序
├── qs_cleaned.csv     # QS 排名数据文件
├── qs_search.py       # 原控制台搜索程序
├── qs_stats.py        # 原控制台统计程序
├── qs_top.py          # 原控制台排名程序
├── requirements.txt   # Python 依赖
├── build.sh           # 构建脚本
├── run.sh             # 启动脚本
└── README.md          # 项目说明

license

MIT License