AI Skill Hub 推荐使用:学术工具MCP 是一款优质的MCP工具。AI 综合评分 7.5 分,在同类工具中表现稳健。如果你正在寻找可靠的MCP工具解决方案,这是一个值得深入了解的选择。
学术工具MCP 是一款遵循 MCP(Model Context Protocol)标准协议的 AI 工具扩展。通过 MCP 协议,它可以让 Claude、Cursor 等主流 AI 客户端直接访问和操作外部工具、数据源和服务,实现 AI 能力的无缝扩展。无论是文件操作、数据库查询还是 API 调用,都可以通过自然语言在 AI 对话中直接触发,极大提升生产效率。
学术工具MCP 是一款遵循 MCP(Model Context Protocol)标准协议的 AI 工具扩展。通过 MCP 协议,它可以让 Claude、Cursor 等主流 AI 客户端直接访问和操作外部工具、数据源和服务,实现 AI 能力的无缝扩展。无论是文件操作、数据库查询还是 API 调用,都可以通过自然语言在 AI 对话中直接触发,极大提升生产效率。
# 方式一:通过 Claude Code CLI 一键安装
claude skill install https://github.com/hunter-heidenreich/academic-tools-mcp
# 方式二:手动配置 claude_desktop_config.json
{
"mcpServers": {
"----mcp": {
"command": "npx",
"args": ["-y", "academic-tools-mcp"]
}
}
}
# 配置文件位置
# macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
# Windows: %APPDATA%/Claude/claude_desktop_config.json
# 安装后在 Claude 对话中直接使用 # 示例: 用户: 请帮我用 学术工具MCP 执行以下任务... Claude: [自动调用 学术工具MCP MCP 工具处理请求] # 查看可用工具列表 # 在 Claude 中输入:"列出所有可用的 MCP 工具"
// claude_desktop_config.json 配置示例
{
"mcpServers": {
"____mcp": {
"command": "npx",
"args": ["-y", "academic-tools-mcp"],
"env": {
// "API_KEY": "your-api-key-here"
}
}
}
}
// 保存后重启 Claude Desktop 生效
An MCP server that gives LLM agents lean, focused tools for working with academic papers. Built on FastMCP.
Look up paper metadata, authors, abstracts, citations, and BibTeX entries. Download and read full paper PDFs section-by-section. Explore reference and citation graphs. Cross-reference with Wikipedia.
Requires Python 3.11+ and uv.
git clone https://github.com/hunter-heidenreich/academic-tools-mcp.git
cd academic-tools-mcp
uv sync
cp .env.example .env # then edit .env with your values
python -m venv ~/.venvs/mineru
source ~/.venvs/mineru/bin/activate
pip install mineru
Then in .env:
PDF_CONVERTER=mineru
PDF_CONVERTER_VENV=~/.venvs/mineru
All configuration is via environment variables in .env. Nothing is required to get started, but some variables unlock higher rate limits.
| Variable | Required | Description |
|---|---|---|
OPENALEX_API_KEY | No | Free API key from [openalex.org](https://openalex.org/settings/api) |
OPENALEX_MAILTO | No | Your email — gets you into the [polite pool](https://docs.openalex.org/how-to-use-the-api/rate-limits-and-authentication#the-polite-pool) (faster) |
CROSSREF_MAILTO | No | Your email — gets you into the Crossref polite pool (10 req/sec vs 5) |
WIKIPEDIA_MAILTO | No | Your email — required by [Wikimedia policy](https://meta.wikimedia.org/wiki/User-Agent_policy) for the User-Agent header |
PDF_CONVERTER | No | PDF-to-markdown backend: mineru (default), marker, or a custom command (see [PDF Pipeline](#pdf-pipeline)) |
PDF_CONVERTER_VENV | No | Path to a virtualenv to activate before running the converter (e.g. ~/.venvs/mineru) |
PDF_CONVERT_TIMEOUT | No | Hard timeout for a single PDF→markdown conversion in seconds (default 1800 = 30 min). Set to none / off / disabled to disable. |
| Tool | Description |
|---|---|
get_paper_references_count | Survey outgoing-reference coverage across both Crossref and OpenCitations in one call — returns per-source counts so you can pick which to page through |
get_paper_references | Paginated outgoing references. Default source="auto" surveys both Crossref and OpenCitations in parallel and pages from whichever has more; pass source="crossref" for structured metadata or source="opencitations" for broader DOI coverage to skip the survey |
get_paper_citations_count | Number of incoming citations (OpenCitations) |
get_paper_citations | Paginated incoming citations with DOIs, dates, self-citation flags, and cross-referenced IDs (OpenCitations) |
search_crossref_by_title | DOI discovery by bibliographic query (also works for bioRxiv papers); each hit warms the works cache so a follow-up get_paper_metadata(doi) is free |
For citations, follow the count-then-page pattern: call get_paper_citations_count first to see the total, then page through with page and page_size. For references the source="auto" default does the survey for you on the first call. Paginated responses include _source (on references) and has_more so agents know which shape to expect and when to stop. This prevents token blowouts on papers with long bibliographies or many citations.
Source trade-off for references: Crossref returns structured reference metadata (author, title, year, journal, DOI) when publishers deposit it; quality varies. OpenCitations aggregates from Crossref, PubMed, DataCite, OpenAIRE, and JaLC — it may have entries Crossref lacks, but returns DOI-to-DOI links only (no bibliographic metadata).
| Tool | Description |
|---|---|
download_pdf | Download and cache the PDF — auto-detects arXiv, ACL Anthology, bioRxiv/medRxiv. Streams chunks to disk (peak memory = 64 KiB) and aborts mid-stream if the response would exceed MAX_PDF_BYTES (default 200 MB). Re-downloading with force_refresh=True cascades: the cached markdown + section index are dropped automatically so the next convert_paper picks up the new bytes. |
convert_paper | Convert PDF to markdown, parse into sections (slow: tens of minutes; PDF_CONVERT_TIMEOUT caps it at 30 min by default). The server runs at most one conversion at a time across all callers — a second concurrent caller gets {busy: True, retryable: True, in_progress: {...}} immediately rather than queueing |
get_paper_sections | Section index with titles, sub-heading previews, token counts |
get_paper_section | Markdown of a section (by index or title substring); truncated by default (16000 chars) |
find_in_paper | Substring (or whole-word) search inside one converted paper. Returns each hit's section + char offset + ~120-char snippet. Char offsets align with get_paper_section's stripped text so you can chain straight to the surrounding context. |
All four tools accept any identifier (arXiv ID, DOI, or freeform label) and auto-route to the correct provider's cache namespace. For papers not hosted on arXiv/ACL/bioRxiv, fetch the PDF yourself and hand it to import_paper — see Manual import below.
The PDF-to-markdown pipeline converts downloaded PDFs into section-level markdown that agents can read piece by piece, avoiding token blowouts from dumping entire papers into context.
The pipeline is converter-agnostic. Set PDF_CONVERTER in .env to choose your backend:
```bash
高质量开源MCP工具,辅助学术研究
AI Skill Hub 为第三方内容聚合平台,本页面信息基于公开数据整理,不对工具功能和质量作任何法律背书。
建议在沙箱或测试环境中充分验证后,再部署至生产环境,并做好必要的安全评估。
✅ MIT 协议 — 最宽松的开源协议之一,可自由商用、修改、分发,仅需保留版权声明。
总体来看,学术工具MCP 是一款质量良好的MCP工具,在同类工具中具备一定竞争力。AI Skill Hub 将持续追踪其更新动态,建议收藏备用,结合自身场景选择合适时机引入使用。
| 原始名称 | academic-tools-mcp |
| 原始描述 | 开源MCP工具:MCP server giving LLM agents lean, identifier-routed tools to look up, read, and。⭐5 · Python |
| Topics | mcpacademic-researchai-agentsarxivbibtexbiorxiv |
| GitHub | https://github.com/hunter-heidenreich/academic-tools-mcp |
| License | MIT |
| 语言 | Python |
收录时间:2026-05-31 · 更新时间:2026-06-01 · License:MIT · AI Skill Hub 不对第三方内容的准确性作法律背书。
选择 Agent 类型,复制安装指令后粘贴到对应客户端