Paper Search MCP Server

纸张查找MCP服务器

A. 模型上下文协议（MCP） 用于从多个平台搜索和下载学术论文的服务器。专为Claude Desktop、Cursor等LLM工具而设计。

中文文档

______________________________________________________________________

支持的平台

核心平台

RePEc/IDEAS功能

RePEc是最大的开放经济学参考书目，具有丰富的搜索选项：

搜索字段：全文/摘要/关键词/标题/作者

排序选项：相关性/最新/最旧/引用/最近和相关

文档类型：期刊文章/工作论文/书籍章节/书籍

机构/期刊筛选器:

______________________________________________________________________

快速开始

安装

从PyPI安装（推荐）：

# Using uv (recommended)
uv pip install paper-find-mcp

# Or using pip
pip install paper-find-mcp

从源代码安装：

# Clone the repository
git clone https://github.com/h-lu/paper-find-mcp.git
cd paper-find-mcp

# Install with uv
uv pip install -e .

# Or use pip
pip install -e .

配置Claude桌面

编辑 ~/Library/Application Support/Claude/claude_desktop_config.json （macOS）：

使用uvx（推荐，无需预安装）：

{
  "mcpServers": {
    "paper_find_server": {
      "command": "uvx",
      "args": ["paper-find-mcp"],
      "env": {
        "SEMANTIC_SCHOLAR_API_KEY": "",
        "CROSSREF_MAILTO": "your_email@example.com",
        "NCBI_API_KEY": "",
        "PAPER_DOWNLOAD_PATH": "~/paper_downloads"
      }
    }
  }
}

使用pip安装包：

{
  "mcpServers": {
    "paper_find_server": {
      "command": "paper-find-mcp",
      "args": [],
      "env": {
        "SEMANTIC_SCHOLAR_API_KEY": "",
        "CROSSREF_MAILTO": "your_email@example.com",
        "NCBI_API_KEY": "",
        "PAPER_DOWNLOAD_PATH": "~/paper_downloads"
      }
    }
  }
}

______________________________________________________________________

使用指南

按学科选择工具

General academic search    → search_semantic or search_crossref
CS/Physics preprints       → search_arxiv
Biomedical                 → search_pubmed + download_scihub(doi)
Economics                  → search_repec (supports NBER, IMF, Fed, AER, etc.)
Cryptography               → search_iacr
Download paywalled papers  → download_scihub(doi) [pre-2023]

典型工作流程

# 1. Search for papers
papers = search_semantic("climate change agriculture", max_results=5)

# 2. Get DOI
doi = papers[0]["doi"]

# 3. Download via Sci-Hub (older papers)
pdf_path = download_scihub(doi)

# 4. Read full text
text = read_scihub_paper(doi)

RePEc经济学搜索示例

# Search NBER working papers
search_repec("inflation expectations", series='nber')

# Search AER journal articles, sorted by newest
search_repec("causal inference", series='aer', sort_by='newest')

# Search Fed papers, with year filter
search_repec("monetary policy", series='fed', year_from=2020)

# Search by author
search_repec("Acemoglu", search_field='author')

# Get paper details (with full abstract)
get_repec_paper("https://ideas.repec.org/p/nbr/nberwo/32000.html")

______________________________________________________________________

完整工具清单

搜索工具

下载工具

阅读工具（PDF→ Markdown）

辅助工具

______________________________________________________________________

环境变量

______________________________________________________________________

发展

# Clone the repository
git clone https://github.com/h-lu/paper-find-mcp.git
cd paper-find-mcp

# Create virtual environment
uv venv && source .venv/bin/activate

# Install dev dependencies
uv pip install -e .

# Run tests
uv run pytest tests/ -v

______________________________________________________________________

许可证

MIT许可证

原始代码基于 论文搜索mcp\ 版权所有（c）2025 OPENAGS

修改和增强\ 版权所有（c）陆海波2025

______________________________________________________________________

🎓 快乐的研究！