@IlyaGusev
基本情報
Academia MCP server: Tools for automatic scientific research
以下の設定を使って、このサーバーを MCP 対応クライアントに追加してください。
{
"mcpServers": {
"academia": {
"command": "python3",
"args": [
"-m",
"academia_mcp",
"--transport",
"stdio"
]
}
}
}ツール一覧はまだありません
この掲載には表示できるツール一覧がまだありません。サーバーにツールがないという意味ではありません。対応機能はプロジェクトのドキュメントをご確認ください。
ツールは README から自動的に抽出されます。メンテナーは ## Tools という見出しの下に記載することで、このタブに反映できます。
Academia MCP is a Model Context Protocol server that provides tools to search, fetch, analyze, and report on scientific papers and datasets. It integrates with ArXiv, ACL Anthology, Hugging Face Datasets, Semantic Scholar, web search engines (Exa, Brave, Tavily), and LaTeX compilation. Designed for researchers and developers who want AI-assisted literature discovery and analysis.
Install via pip3 install academia-mcp or use uv for development. Run the server with python -m academia_mcp --transport stdio for local MCP clients or --transport streamable-http for HTTP. Configure optional environment variables to enable specific tools (e.g., OPENROUTER_API_KEY for LLM tools, WORKSPACE_DIR for file operations).
Python 3.12 or higher is required. For LaTeX/PDF tools, a LaTeX distribution with `pdflatex` and `latexmk` on PATH is needed (e.g., install via `sudo apt install texlive-latex-base texlive-fonts-recommended texlive-latex-extra texlive-science latexmk`).
Set the `ENABLE_AUTH` environment variable to `true` and optionally provide a `TOKENS_FILE` path. Use the `academia_mcp auth` commands to issue, list, and revoke tokens. Include the token in the `Authorization: Bearer <token>` header or as the `apiKey` query parameter.
Set `OPENROUTER_API_KEY` for LLM tools (`document_qa`, `review_pdf_paper`, bitflip tools). Set `WORKSPACE_DIR` for `compile_latex`, `read_pdf`, `download_pdf_paper`, and `review_pdf_paper`. Set one of `EXA_API_KEY`, `BRAVE_API_KEY`, or `TAVILY_API_KEY` for web search tools.
Build the image with `docker build -t academia_mcp .` or use the existing image `phoenix120/academia_mcp`. Run with `docker run -p 5056:5056 -e PORT=5056 -e OPENROUTER_API_KEY=... -e WORKSPACE_DIR=/workspace -v "$PWD/workdir:/workspace" academia_mcp`.
The server supports three transports: `stdio` (for local MCP clients like Claude Desktop), `sse`, and `streamable-http`. Use the `--transport` flag to select one. The `host`/`port` options apply only to HTTP transports; default port is `5056` (or from `PORT` environment variable).
Model Context Protocol Servers
This open-source curriculum introduces the fundamentals of Model Context Protocol (MCP) through real-world, cross-language examples in .NET, Java, TypeScript, JavaScript, Rust and Python. Designed for developers, it focuses on practical techniques for building modular, scalable,
A free remote MCP server for The Living Bread. Bible passages read word for word from a stored public domain text with an evidence label, keyword search and cross references, real churches, gatherings and prayer groups w
Yuppsy MCP: build and publish real estate websites from your AI assistant. Yuppsy lets your AI assistant create live websites for real estate agents and individual property listings. You don't need a designer or any code
See and control a real Windows PC you own — from any MCP client, locally or remotely. Observe (UIA + screenshots), click/type/drag/scroll, launch apps, owner Live View. BYOH: your machine, your key.
コメント