Embedding MCP Server

What is kb-mcp-server?

The kb-mcp-server is an implementation of a Model Context Protocol (MCP) server that allows users to build and serve knowledge bases using semantic search and AI-driven text processing.

How to use kb-mcp-server?

To use the kb-mcp-server, you can build a knowledge base using the kb_builder command-line tool or directly through the txtai programming interface. Once built, you can start the MCP server to serve the knowledge base.

Key features of kb-mcp-server?

• Unified vector database for semantic search and knowledge graph capabilities.
• Ability to build and manage knowledge bases from various document sources.
• Support for exporting knowledge bases as compressed archives (.tar.gz).
• Local-first architecture allowing operation without external data sharing.

Use cases of kb-mcp-server?

1. Creating a searchable knowledge base for research documents.
2. Building a knowledge graph from technical documentation.
3. Serving AI-driven text processing for various applications.

FAQ from kb-mcp-server?

• Can I use my own documents to build a knowledge base?

Yes! You can process documents from various sources to create your knowledge base.

• Is there a command-line interface available?

Yes! The kb_builder module provides a command-line interface for building and managing knowledge bases.

• How do I start the MCP server?

You can start the MCP server using the command kb-mcp-server --embeddings /path/to/knowledge_base.