Academix

Academic MCP Server

A Model Context Protocol (MCP) server for searching, downloading, and analyzing academic papers from multiple sources.

Aggregates data from OpenAlex, DBLP, Semantic Scholar, arXiv, and CrossRef to provide a unified academic research interface for LLMs.

Designed for seamless integration with large language models like Claude Desktop and OpenCode.

Features

• 🔍 Multi-Source Search: Unified search across OpenAlex (100K+ free calls/day), DBLP (CS papers), Semantic Scholar (AI recommendations), arXiv (preprints), and CrossRef (DOI resolution).
• 🧠 Smart ID Resolution: Automatically detects and handles DOIs, arXiv IDs, OpenAlex IDs, Semantic Scholar IDs, and DBLP keys.
• 📚 BibTeX Export: Native BibTeX support from DBLP (high quality) with automatic fallback generation for other sources. Supports batch export.
• 📊 Citation Analysis: Retrieve citation counts, citing papers, and generate citation network data for visualization.
• 🤖 AI Recommendations: Leverage Semantic Scholar's AI engine to find related papers based on content and citations.
• ⚡ Asynchronous: Built with httpx and asyncio for high-performance concurrent API requests.

Installation

Quick Start (for Users)

If you are using uv (recommended):

uv tool install academix

Or with pip:

pip install academix

Development Setup (for Contributors)

1. Clone the repository:

``bash git clone https://github.com/your-org/academix.git cd academix ``

1. Install dependencies:

``bash uv sync ``

1. Run the server:

``bash uv run academix ``

Configuration

1. Claude Desktop Integration

To use this server with Claude Desktop, add the following to your configuration file:

macOS: ~/Library/Application Support/Claude/claude_desktop_config.json Windows: %APPDATA%\Claude\claude_desktop_config.json

{
  "mcpServers": {
    "academix": {
      "command": "uv",
      "args": [
        "run",
        "--directory",
        "/absolute/path/to/academix", 
        "academix"
      ],
      "env": {
        "ACADEMIX_EMAIL": "your.email@example.com",
        "SEMANTIC_SCHOLAR_API_KEY": "optional_api_key"
      }
    }
  }
}

> Note: Replace /absolute/path/to/academix with the actual path where you cloned the repository. Setting ACADEMIX_EMAIL is highly recommended to access the "polite pool" for OpenAlex and CrossRef (higher rate limits).

2. OpenCode Integration

For OpenCode, modify your configuration file (usually at ~/.config/opencode/opencode.json):

{
  "$schema": "https://opencode.ai/config.json",
  "mcp": {
    "academix": {
      "type": "local",
      "command": [
        "uv",
        "run",
        "--directory",
        "/absolute/path/to/academix",
        "academix"
      ],
      "environment": {
        "ACADEMIX_EMAIL": "your.email@example.com",
        "SEMANTIC_SCHOLAR_API_KEY": "optional_api_key"
      },
      "enabled": true,
      "timeout": 10000
    }
  }
}

Configuration Details:

• type: Must be "local".
• command: The full command array to run the server. Use uv run pointing to the directory.
• directory: Replace /absolute/path/to/academix with your actual path.
• environment: Add API keys or email here. ACADEMIX_EMAIL is recommended.
• timeout: Optional. Increased to 10000 (10s) to handle network latency.

Data Source Coverage

Feature Support Matrix

| Feature | OpenAlex | DBLP | CrossRef | Semantic Scholar | arXiv | |---------|:--------:|:----:|:--------:|:----------------:|:-----:| | Paper Search | ✅ | ✅ | ✅ | ✅ | ✅ | | Get Paper Details | ✅ | ✅ | ✅ | ✅ | ✅ | | BibTeX Export | ✅ | ✅ Native | ✅ | ✅ | ✅ | | Citation Retrieval | ✅ | ❌ | ❌ | ✅ | ❌ | | Author Search | ✅ | ✅ | ✅ | ✅ | ✅ | | Related Papers (AI) | ❌ | ❌ | ❌ | ✅ | ❌ |

BibTeX Quality by Source

| Source | Quality | Notes | |--------|---------|-------| | DBLP | ⭐⭐⭐ Excellent | Native BibTeX export. Best for CS papers. Includes complete venue info. | | CrossRef | ⭐⭐ Good | Accurate metadata from publishers. Requires valid DOI. | | arXiv | ⭐⭐ Good | Correct preprint info. Entry type is @misc with eprint field. | | Semantic Scholar | ⭐⭐ Good | Generated from metadata. May lack venue details. | | OpenAlex | ⭐ Variable | Generated from metadata. Some fields may be incomplete. |

Recommendation: For CS papers, prioritize DBLP for BibTeX. For papers with DOI, CrossRef provides reliable metadata.

MCP Tools

| Tool | Description | |------|-------------| | academic_search_papers | Search papers by keywords, title, author, DOI, date, venue. Supports sorting by relevance, date, or citations. | | academic_get_paper_details | Get full metadata for a paper using any supported ID format. | | academic_get_bibtex | Export BibTeX citations (single or batch). Prioritizes DBLP for high-quality metadata. | | academic_get_citations | Get papers that cite a given paper (via OpenAlex). | | academic_search_author | Find all papers by an author name. | | academic_get_related_papers | AI-powered related paper recommendations (via Semantic Scholar). | | academic_get_citation_network | Get citation network data (nodes/edges) for visualization. | | academic_cache_stats | View cache hit rates and statistics. |

Usage Examples

Search for Papers

Find papers about "LLM agents" sorted by citation count:

use academic_search_papers with:
{
  "query": "LLM agents",
  "limit": 5,
  "sort": "citation_count"
}

Get BibTeX for a List of Papers

Generate BibTeX for multiple papers (mixed ID formats supported):

use academic_get_bibtex with:
{
  "paper_ids": "10.1038/nature12345, 10.48550/arXiv.2310.08560"
}

Analyze Citations

Find who is citing a specific paper:

use academic_get_citations with:
{
  "paper_id": "W2741809807" // OpenAlex ID or DOI
}

API Sources & Rate Limits

| Source | Rate Limits | Authentication | Best For | |--------|-------------|----------------|----------| | OpenAlex | 100K/day (with email) | Email (optional) | General search, citations, author data | | DBLP | Reasonable use | None | CS papers, high-quality BibTeX | | Semantic Scholar | 100/5min (higher with key) | API key (optional) | AI Recommendations | | arXiv | Unlimited (polite) | None | Preprints (CS, Math, Physics) | | CrossRef | Dynamic | Email (optional) | DOI resolution |

Development

Run Tests

# Run all tests
uv run pytest

# Run with coverage report
uv run pytest --cov=academix

Project Structure

academix/
├── src/academix/
│   ├── server.py          # MCP server entry point
│   ├── aggregator.py      # Orchestrator for multiple API clients
│   ├── clients/           # Individual API client implementations
│   │   ├── openalex.py
│   │   ├── dblp.py
│   │   ├── semantic.py
│   │   ├── arxiv_client.py
│   │   └── crossref.py
│   └── models.py          # Pydantic data models
├── tests/                 # Unit and integration tests
├── pyproject.toml         # Dependencies and config
└── README.md

License

MIT License

Source & license

This open-source MCP server is cataloged on AgentStack and links to its original source — we do not rehost the code.

• Author: xingyulu23
• Source: xingyulu23/Academix
• License: MIT

Install and usage instructions live in the source repository linked above.