🧠 NexusMind

What is NexusMind?

NexusMind is an intelligent scientific reasoning framework that leverages a Neo4j graph database and the Graph-of-Thoughts approach to perform complex research tasks. It implements the Model Context Protocol (MCP) to integrate with AI applications such as Claude Desktop, providing an Advanced Scientific Reasoning Graph-of-Thoughts (ASR-GoT) engine for researchers and developers.

How to use NexusMind?

Clone the repository, install dependencies with poetry install, configure config/settings.yaml, then run the development server via uvicorn asr_got_reimagined.main:app --reload at localhost:8000. For Docker deployment, use docker-compose up --build. The MCP endpoint is POST /mcp and health check is GET /health.

Key features of NexusMind

• Graph-based scientific reasoning
• Dynamic confidence scoring with multi-dimensional evaluations
• Neo4j graph database integration
• Model Context Protocol (MCP) support
• Dockerized for easy deployment
• Modular and extensible design

Use cases of NexusMind

• Analyze relationships between microbiome diversity and cancer progression.
• Process complex scientific queries using graph-based reasoning chains.
• Integrate with Claude Desktop for interactive research assistance.
• Deploy a self-contained reasoning engine via Docker Compose.
• Explore and iterate on scientific hypotheses via MCP clients.

FAQ from NexusMind

What is NexusMind and how does it differ from other AI reasoning tools?

NexusMind is an MCP server that uses a Neo4j graph database and Graph-of-Thoughts for scientific reasoning, unlike standard LLM-based tools. It implements the Model Context Protocol for integration with Claude Desktop and other MCP clients.

What are the prerequisites to run NexusMind?

Python 3.11+, Poetry, Docker and Docker Compose (optional), and a running Neo4j instance with the APOC library installed. The provided Docker Compose files include Neo4j with APOC pre-configured.

How do I deploy NexusMind with Docker?

Run docker-compose up --build for development or docker-compose -f docker-compose.prod.yml up --build -d for production. The API will be available at http://localhost:8000, with Swagger docs at /docs.

Does NexusMind support persistent conversational sessions?

Currently, each query is processed independently; the session_id identifies a single query-response cycle but does not persist graph state between calls. Persistent sessions are a planned future enhancement.

What transports and authentication does NexusMind use?

Transport is HTTP via FastAPI. The MCP endpoint is POST /mcp. Authentication details are not specified in the README; configuration is handled through settings.yaml or environment variables.