通过MCP工作流搜索和分析arXiv论文
Connect an MCP-capable assistant to arXiv search, paper download, local paper storage, and research prompts for grounded literature review workflows.
Prerequisites
Python 3.11+, uv or supported MCP bundle installer, MCP-compatible client such as Claude Desktop, VS Code, Kiro, or another stdio MCP host
Installation
Use the upstream install or setup path that matches your environment:
- uv pip install -e ".[pro]"
Requirements and caveats from upstream:
-
- python
- Read the full text of a locally downloaded paper in markdown. Requires download_paper to be called first. Use start and max_chars with the returned next_start value to page through large papers.
Basic usage or getting-started notes:
- example, instructing it to exfiltrate data, invoke other tools with unintended
- asks you to run commands or visit external URLs that were not part of your
- Run the test suite:
- Source: https://github.com/1991513ccie-png/skills
- Extracted from upstream docs: https://raw.githubusercontent.com/blazickjp/arxiv-mcp-server/HEAD/README.md
Documentation
- https://github.com/1991513ccie-png/skills
