Huggingface Daily Papers

Name: Huggingface Daily Papers
Author: huangxinping

@huangxinping

Repository

About Huggingface Daily Papers

A MCP (Model Context Protocol) server for fetching HuggingFace daily papers.

#huggingface #paper

Basic information

Config

Add this server to your MCP-compatible client using the configuration below.

{
  "mcpServers": {
    "huggingface-papers": {
      "command": "uvx",
      "args": [
        "huggingface-daily-paper-mcp"
      ]
    }
  }
}

Tools

Get HuggingFace daily papers for a specific date

Get today's HuggingFace daily papers

Get yesterday's HuggingFace daily papers

Overview

What is Huggingface Daily Papers?

Huggingface Daily Papers is a Model Context Protocol (MCP) server that fetches daily papers from Hugging Face. It provides tools and resources to retrieve papers for today, yesterday, or a specific date, including title, authors, abstract, tags, votes, and download links. The server is intended for AI developers and agents using MCP-compatible clients (e.g., Claude Code, Cursor IDE) to stay current with machine learning research.

How to use Huggingface Daily Papers?

Install and run directly using uvx huggingface-daily-paper-mcp (recommended) or clone the repository for local development. The server runs as an MCP server via python main.py. Configuration is done by adding the server to your MCP client settings (e.g., in Claude Code, Cursor, Windsurf, or VS Code Continue) with the command uvx and argument ["huggingface-daily-paper-mcp"].

Key features of Huggingface Daily Papers

Fetch papers for today, yesterday, or any date (YYYY-MM-DD).
Returns title, authors, abstract, tags, votes, and submitter.
Includes paper URL and PDF download link.
Exposes MCP tools (get_papers_by_date, get_today_papers, get_yesterday_papers).
Provides MCP resources (papers://today, papers://yesterday).
ArXiv integration for complete author lists.

Use cases of Huggingface Daily Papers

Stay updated with the latest Hugging Face daily papers.
Integrate paper fetching into AI coding assistants like Claude Code or Cursor.
Automate retrieval and analysis of recent machine learning research.
Build custom dashboards or notification systems for specific paper tags.

FAQ from Huggingface Daily Papers

What tools does the server provide?

Three tools: get_papers_by_date (with a date parameter in YYYY-MM-DD format), get_today_papers, and get_yesterday_papers (both with no parameters).

What resources does the server expose?

Two resources: papers://today and papers://yesterday, returning today's and yesterday's papers as JSON data.

What are the runtime requirements?

Python 3.10+ and dependencies: MCP, Requests, BeautifulSoup4, pytest, and uv (package manager). The server uses uvx for direct execution.

How do I install the server?

Run uvx huggingface-daily-paper-mcp in your terminal. For local development, clone the repository and use uv sync.

How do I configure the server for Claude Code?

Add to your MCP configuration under mcpServers with "command": "uvx" and "args": ["huggingface-daily-paper-mcp"].

Comments

More AI & Agents MCP servers

Model Context Protocol for Unreal Engine

chongdashu

Enable AI assistant clients like Cursor, Windsurf and Claude Desktop to control Unreal Engine through natural language using the Model Context Protocol (MCP).

2.0K