What is KuzuMem-MCP?

KuzuMem-MCP is a TypeScript implementation of a distributed memory bank designed for code agents. It leverages KuzuDB to store memories in a graph database, enabling repository and branch filtering capabilities, and adheres to the Model Context Protocol (MCP).

What is KuzuDB?

KuzuDB is a high-performance, in-memory graph database used by KuzuMem-MCP for efficient storage and retrieval of graph-structured data, ideal for representing relationships between entities in software projects.

What is MCP (Model Context Protocol)?

MCP is an open protocol that standardizes how applications provide context to Large Language Models (LLMs), facilitating interaction with external data sources and tools.

What are the key features of KuzuMem-MCP?

Key features include a unified tool architecture, thread-safe singleton pattern, distributed graph structure, repository & branch awareness, asynchronous operations, multiple access interfaces, KuzuDB backend, full MCP compliance, progressive results streaming, and client project root isolation.

What unified tools are included in KuzuMem-MCP?

The system broadcasts 10 unified tools: memory-bank, entity, introspect, context, query, associate, analyze, detect, bulk-import, and search, covering all memory bank operations.

How do I install KuzuMem-MCP?

Clone the repository, install dependencies using npm install, and build the project using npm run build.

What configuration is required for KuzuMem-MCP?

Create a .env file with database and server configurations, including DB_FILENAME, HTTP_STREAM_PORT, HOST, and DEBUG. Also, add the server configuration to your IDE’s MCP settings.

What is the Agent Development Loop in KuzuMem-MCP?

It’s a five-phase loop (ANALYZE, BLUEPRINT, CONSTRUCT, VALIDATE, ROLLBACK) that ensures consistency and adherence to governance rules for AI agent development.

What kind of license does KuzuMem-MCP have?

KuzuMem-MCP is licensed under Apache-2.0.

Can I contribute to KuzuMem-MCP?

Yes, contributions are welcome. Ensure all tests pass, code follows existing style, new features include tests, and documentation is updated.

What are some future improvements planned for KuzuMem-MCP?

Future improvements include full-text search, vector embeddings, advanced graph algorithms, graph schema updates, and complete semantic search.