CiteNexus MCP
Enables AI agents to access academic citations, fetch metadata, and generate BibTeX entries directly from Google Scholar via MCP tools.
README
CiteNexus MCP
The AI-Native Academic Citation & Research Engine
CiteNexus is a fast, lightweight Model Context Protocol (MCP) server designed to give AI agents (like Claude Code, Gemini CLI, Cursor, and Windsurf) native access to the global academic literature graph.
The name says it all: it is the nexus where your AI assistant connects directly to the world's academic citations. Rather than forcing researchers to break their writing flow to navigate web interfaces, CiteNexus brings powerful reference management, citation generation, and metadata formatting directly into the IDEs and terminals where the writing actually happens.
🌟 The Vision
For decades, reference management has meant opening dedicated desktop applications (like Zotero or EndNote) or navigating browser-based walled gardens. While modern tools like Google Scholar Labs offer fantastic human-centric reading experiences, they remain isolated from the actual writing environment.
CiteNexus takes a different approach: Innovation through Integration.
We believe the future of research is Agentic. Your AI assistant should be able to seamlessly fetch, format, and verify citations without you ever leaving your editor. CiteNexus acts as the critical bridge between massive academic databases (like Google Scholar) and your local AI workflows. It doesn't compete with the giants of academic search; it unlocks their full potential for the AI era.
Why CiteNexus?
- Cluster-First Architecture: Uses Google Scholar "Cluster IDs" as the universal source of truth, bypassing the brittleness of DOIs, mismatched titles, or broken URLs.
- LLM-Powered Formatting (Elicitation): Replaces thousands of lines of fragile parsing code with dynamic AI formatting. Need a bespoke BibTeX format for a niche IEEE conference? CiteNexus handles it gracefully.
- Where You Write: Integrates directly into your AI coding assistants, meta-prompting frameworks (like WTF-P), and terminal agents. No more tab-switching.
🛠️ Core Capabilities
CiteNexus exposes four focused MCP tools to your AI agent:
find-scholar-id: Converts any messy input (a loose title, an ArXiv ID, a DOI, or a fragmented citation) into a universal, stable Google Scholar Cluster ID.get-citation: Fetches the complete metadata for a Cluster ID and generates a perfectly accurate BibTeX entry.enhance-citation: Applies custom templates, rules, and notes to an existing citation (e.g., "Change to first-initial only", "Add a custom note regarding methodology").paper-metrics: Retrieves impact analytics, citation counts, and top related papers to help your agent evaluate a source's significance.
🚀 Quickstart
CiteNexus is packaged with uv for lightning-fast installation and execution.
Prerequisites
You need a SerpAPI Key to query Google Scholar.
export SERP_API_KEY="your-serpapi-key"
(Optional) Fallback API Configuration: If your primary MCP client (e.g., Claude Code, Gemini CLI) does not yet support native MCP Elicitation, CiteNexus will automatically fall back to an OpenAI-compatible API to perform its data formatting.
export OPENAI_API_KEY="sk-..."
# Optional overrides for local models (Ollama, vLLM, etc.):
# export OPENAI_API_BASE="http://localhost:11434/v1"
# export OPENAI_MODEL="llama3"
Running via uvx
You can run the server instantly without permanently installing it into your global environment:
uvx cite-nexus-mcp
IDE / Agent Integration Examples
Claude Desktop
Add CiteNexus to your claude_desktop_config.json:
{
"mcpServers": {
"cite-nexus": {
"command": "uvx",
"args": ["cite-nexus-mcp"],
"env": {
"SERP_API_KEY": "your-serp-api-key",
"OPENAI_API_KEY": "your-openai-api-key-if-needed"
}
}
}
}
Cursor / Windsurf
Provide the exact same command (uvx cite-nexus-mcp) and environment variables in the MCP configuration panel of your IDE settings.
🏗️ Development
To build on top of CiteNexus or run it locally:
- Install
uv:curl -LsSf https://astral.sh/uv/install.sh | sh - Clone and install:
git clone https://github.com/akougkas/cite-nexus-mcp.git cd cite-nexus-mcp uv sync - Run the development server:
uv run cite-nexus-mcp
🤝 Philosophy & The Future
CiteNexus is built on the "Engine in the Car" philosophy. It is designed to be the ultimate citation engine that powers larger, more ambitious academic AI frameworks. As the academic ecosystem evolves, CiteNexus will grow to encompass citation graph traversal, local library (PDF) syncing, and hallucination verification, empowering researchers to do their best work at the speed of thought.
📄 License
MIT License
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.