kb-mcp-server
26
Embedding MCP Server is a Model Context Protocol server powered by txtai, offering semantic search and AI-driven text processing with a standard interface. It features comprehensive knowledge graph capabilities and advanced causal boosting to enhance search relevance. The server can be configured and deployed easily using various installation methods.
Embedding MCP Server
A Model Context Protocol (MCP) server implementation powered by txtai, providing semantic search, knowledge graph capabilities, and AI-driven text processing through a standardized interface.
The Power of txtai: All-in-one Embeddings Database
- Unified Vector Database
- Semantic Search
- Knowledge Graph Integration
- Portable Knowledge Bases
- Extensible Pipeline System
- Local-first Architecture
How It Works
- Knowledge base builder tool for creating and managing bases
- MCP server provides interface to access the knowledge base
Installation
Recommended: Using uv with Python 3.10+
- Install and set up a virtual environment
Using conda
- Create a new conda environment for installation
From Source
- Clone the repository and install dependencies
Command Line Usage
Building a Knowledge Base
- Use command-line or Python module for building, updating, and searching knowledge bases
Starting the MCP Server
- Use various methods to start the server, such as PyPI installed command or Python module
MCP Server Configuration
- Configured using environment variables or command-line arguments
Advanced Features
Knowledge Graph Capabilities
- Automatic graph construction, traversal, path finding, and community detection
Causal Boosting Mechanism
- Enhances search relevance by identifying causal relationships