aminer

AMiner MCP 服务器 exposes a set of paper search and analysis capabilities via MCP (Model Context Protocol). It supports keyword, venue, and author searches, as well as an advanced multi-criteria search. The server also includes an AI-assisted paper search prompt template to guide research tasks. You can interact with these tools through MCP-compatible clients by calling the provided RPC methods (search_papers_by_keyword, search_papers_by_venue, search_papers_by_author, search_papers_advanced, and paper_search_assistant). Each tool supports pagination and ordering options, and the results include detailed paper information formatted for consumption by MCP clients.

Prerequisites:

• Node.js 18+ installed on your machine
• Access to npm or pnpm (pnpm is used by the project in its scripts)
• An AMiner API key (to be provided via environment variable)

Installation steps:

1. Install Node.js 18+ from the official site if not already installed.

2. Ensure you have an AMiner API key. Create or obtain one from the AMiner API portal.

3. Install and run the MCP server via npx (no local install required):

   export AMINER_API_KEY=YOUR_AMINER_API_KEY npx -y @scipen/aminer-mcp-server

4. Alternatively, if you want to install locally for development, clone the repository and install dependencies:

   git clone https://github.com/scipenai/aminer-mcp-server.git cd aminer-mcp-server pnpm install export AMINER_API_KEY=YOUR_AMINER_API_KEY pnpm start

5. Configure your MCP client to point to the server as shown in the example configuration.

Tips and notes:

• Set the AMINER_API_KEY environment variable before starting the server to enable access to AMiner data.
• The server supports pagination (page) and page size (size) for all search operations. Respect the documented max sizes (default 10, max 10 for most endpoints).
• The search_papers_advanced endpoint requires at least one of keyword, venue, or author to be provided.
• Use the paper_search_assistant tool to generate AI-assisted prompts for literature exploration, using fields like research_topic and optional search_focus (recent, highly_cited, or comprehensive).
• If you encounter API rate limits or connectivity issues, verify your API key, network access, and ensure you’re using a supported Node.js runtime (18+).
• The MCP protocol uses JSON-RPC 2.0; ensure your client is compatible and request/response payloads match the documented shapes in the README.