Haiku RAG is an intelligent retrieval - augmented generation system built on LanceDB, Pydantic AI, and Docling. It supports hybrid search, re - ranking, Q&A agents, multi - agent research processes, and provides local - first document processing and MCP server integration.

A model context protocol server based on LanceDB, providing functions such as vector storage, similarity search, and metadata management

An MCP server implementation based on the LanceDB vector database, providing semantic memory storage and retrieval functions.

Context Lens is a local semantic search tool that can convert any content into a searchable knowledge base, enabling AI assistants to understand the meaning rather than just match keywords. It uses a built - in LanceDB vector database, supports local files, GitHub repositories, and URL content, does not require an API key or cloud service, and processes data completely locally.

The MCP Document Indexer is a Python - based local document indexing and search server that uses the LanceDB vector database and a local LLM (through Ollama) to implement real - time monitoring, multi - format document processing, and semantic search, and provides tools for AI assistants such as Claude through the Model Context Protocol (MCP).

An MCP server based on LanceDB that enables LLMs to directly interact with local documents through proxy RAG and hybrid search.

A Node.js-based vector search project that uses the LanceDB database and Ollama embedding model to implement document similarity search functionality

Node.js vector search implementation based on LanceDB and Ollama

A serverless MCP service based on LanceDB, providing document storage, retrieval, and table information query functions

LanceDB server project for MCP service

An MCP server project for providing query services for the SQLite database and LanceDB vector database of Chicago Public Schools and their communities