Huggingface Daily Papers
A MCP (Model Context Protocol) server for fetching HuggingFace daily papers.
README
HuggingFace Daily Papers MCP Server
A MCP (Model Context Protocol) server for fetching HuggingFace daily papers.
Features
- Fetch today's, yesterday's or specific date HuggingFace papers
- Provides paper title, authors, abstract, tags, votes, and submitted by info
- Includes paper links and PDF download links
- Supports MCP tools and resource interfaces
- ArXiv integration for complete author lists
- Complete error handling and logging
- Comprehensive test coverage
Installation & Usage
Option 1: Direct execution with uvx (Recommended)
Install and run directly using uvx:
uvx huggingface-daily-paper-mcp
This will automatically install the package and its dependencies, then start the MCP server.
Option 2: Local development
For local development, clone the repository and install dependencies:
git clone https://github.com/huangxinping/huggingface-daily-paper-mcp.git
cd huggingface-daily-paper-mcp
uv sync
Local usage commands
Run as MCP Server (for development):
python main.py
Test Scraper Function:
python scraper.py
Run Tests:
uv run -m pytest test_mcp_server.py -v
Build Package:
uv build
MCP Interface
Tools
-
get_papers_by_date
- Description: Get HuggingFace papers for a specific date
- Parameters:
date(YYYY-MM-DD format)
-
get_today_papers
- Description: Get today's HuggingFace papers
- Parameters: None
-
get_yesterday_papers
- Description: Get yesterday's HuggingFace papers
- Parameters: None
Resources
-
papers://today
- Today's papers JSON data
-
papers://yesterday
- Yesterday's papers JSON data
Project Structure
huggingface-daily-paper-mcp/
├── main.py # MCP server main program
├── scraper.py # HuggingFace papers scraper module
├── test_mcp_server.py # MCP server test cases
├── README.md # Project documentation
├── .gitignore # Git ignore file
├── pyproject.toml # Project configuration file
└── uv.lock # Dependency lock file
Tech Stack
- Python 3.10+: Programming language
- MCP: Model Context Protocol framework
- Requests: HTTP request library
- BeautifulSoup4: HTML parsing library
- pytest: Testing framework
- uv: Python package manager
Development Standards
- Use uv native commands for package management
- Follow Python PEP 8 coding standards
- Include type hints and docstrings
- Complete error handling and logging
- Write unit tests to ensure code quality
Example Output
Single paper data structure:
{
"title": "CMPhysBench: A Benchmark for Evaluating Large Language Models in Condensed Matter Physics",
"authors": ["Weida Wang", "Dongchen Huang", "Jiatong Li", "..."],
"abstract": "CMPhysBench evaluates LLMs in condensed matter physics using calculation problems...",
"url": "https://huggingface.co/papers/2508.18124",
"pdf_url": "https://arxiv.org/pdf/2508.18124.pdf",
"votes": 15,
"submitted_by": "researcher123",
"scraped_at": "2025-08-27T10:30:00.123456"
}
MCP Tool output format:
Title: CMPhysBench: A Benchmark for Evaluating Large Language Models in Condensed Matter Physics
Authors: Weida Wang, Dongchen Huang, Jiatong Li, Tengchao Yang, Ziyang Zheng...
Abstract: CMPhysBench evaluates LLMs in condensed matter physics using calculation problems...
URL: https://huggingface.co/papers/2508.18124
PDF: https://arxiv.org/pdf/2508.18124.pdf
Votes: 15
Submitted by: researcher123
--------------------------------------------------
AI IDE/CLI Configuration
Claude Code (CLI)
Add to your MCP configuration:
{
"mcpServers": {
"huggingface-papers": {
"command": "uvx",
"args": ["huggingface-daily-paper-mcp"]
}
}
}
Cursor IDE
Add to your .cursorrules or MCP settings:
{
"mcp": {
"servers": {
"huggingface-papers": {
"command": "uvx",
"args": ["huggingface-daily-paper-mcp"],
"env": {}
}
}
}
}
Windsurf IDE
Add to your Windsurf MCP configuration:
{
"mcpServers": {
"huggingface-papers": {
"command": "uvx",
"args": ["huggingface-daily-paper-mcp"]
}
}
}
VS Code with Continue Extension
Add to your continue configuration:
{
"mcp": {
"servers": {
"huggingface-papers": {
"command": "uvx",
"args": ["huggingface-daily-paper-mcp"]
}
}
}
}
Other MCP-Compatible Tools
For any MCP-compatible client, use:
# Command
uvx huggingface-daily-paper-mcp
# Or with Python path
python -m main
License
MIT License
推荐服务器
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。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。