What is the Model Context Protocol (MCP)?

MCP is an open protocol enabling seamless integration between LLM applications and external data sources and tools, providing a standardized way for LLMs to access context.

It implements 'add-memory' to store text with vector embeddings and 'search-memories' to retrieve semantically similar memories based on a query.

What is LanceDB MCP Server?

It's a Model Context Protocol (MCP) server for LanceDB, an embedded vector database. It allows LLM applications to store and retrieve memories using semantic similarity search.

What embedding provider does LanceDB MCP Server use?

It uses sentence-transformers for embedding generation, specifically the BAAI/bge-small-en-v1.5 model, by default.

How can I integrate LanceDB MCP Server with Claude?

You can configure Claude to use the LanceDB MCP server by modifying the claude_desktop_config.json file with the command to run the server.

LanceDB

Name: LanceDB
Author: kyryl-opens-ml

•

Provides a Model Context Protocol (MCP) server enabling LLM applications to store and retrieve memories using LanceDB, an embedded vector database.

LanceDB is a Model Context Protocol (MCP) server that integrates LanceDB, an embedded vector database, to provide a semantic memory layer for Large Language Model (LLM) applications. It allows you to store text with vector embeddings for later retrieval, enabling LLMs to access external data sources and tools seamlessly. The server implements 'add-memory' and 'search-memories' tools, allowing for the addition of new memories and the retrieval of semantically similar memories based on a given query.

Key Features

01Uses sentence-transformers for embedding generation

02Configurable database path, collection name, and embedding provider

03Implements 'add-memory' tool for storing text with vector embeddings

04Implements 'search-memories' tool for retrieving semantically similar memories

05Supports semantic similarity search using a configurable threshold

065 GitHub stars

Use Cases

01Enhancing chat interfaces with memory retrieval capabilities

02Building AI-powered IDEs with access to contextual information

03Creating custom AI workflows with semantic memory support