arXiv MCP Skill

This skill enables deep research into academic papers hosted on arXiv.

Quick Start

1. •Search for papers: Use keywords to find relevant research.
2. •Get paper details: Fetch metadata, abstracts, and categories for a specific arXiv ID.
3. •Analyze research: Summarize or extract insights from paper abstracts.

Workflow

1. Searching for Research

When a user asks for "latest papers on X" or "research about Y":

• •Use mcporter call arxiv.search or equivalent.
• •Category Search: You can filter by category using cat:category_name.
  • •Example: cat:cs.AI for Artificial Intelligence.
  • •Example: cat:cs.LG for Machine Learning.
• •Format the output as a list with titles, authors, and dates.

2. Deep Dive into a Paper

When a user provides an arXiv ID (e.g., 1706.03762):

• •Use mcporter call arxiv.get_details or equivalent.
• •Present the abstract and key info.

3. Summarization

If the user wants a summary:

• •Fetch the abstract and provide a high-level overview.

MCP Configuration

This skill relies on the arxiv MCP server. Ensure it is configured in mcporter:

{
  "command": "npx -y arxiv-mcp-server"
}

Related Skills

• •summarize: Use for long-form analysis if the paper text is available via URL.
• •browser: Use to download PDFs or view papers on the arXiv website.