LanceDB Vector Search MCP server

This Node.js server provides vector search capabilities using LanceDB with Ollama's embedding model. It enables you to perform semantic search on documents through a Model Context Protocol (MCP) service that integrates with Claude Desktop.

Prerequisites

• Node.js (v14 or later)
• Ollama running locally with the nomic-embed-text model
• LanceDB storage location with read/write permissions

Installation

Install the required dependencies:

pnpm install

Basic Usage

You can run the vector search test script to verify functionality:

pnpm test-vector-search

Or execute it directly:

node test-vector-search.js

MCP Server Configuration

To integrate with Claude Desktop as an MCP service, add the following to your MCP configuration JSON:

{
  "mcpServers": {
    "lanceDB": {
      "command": "node",
      "args": [
        "/path/to/lancedb-node/dist/index.js",
        "--db-path",
        "/path/to/your/lancedb/storage"
      ]
    }
  }
}

Be sure to replace:

• /path/to/lancedb-node/dist/index.js with the path to the compiled index.js file
• /path/to/your/lancedb/storage with the path to your LanceDB storage directory

Vector Search Functionality

How It Works

The server:

1. Connects to a LanceDB database at the configured path
2. Uses Ollama's API (at http://localhost:11434/api/embeddings) to generate embeddings
3. Performs vector similarity searches against stored documents
4. Returns the most relevant results with similarity scores

Example Usage

The included example searches for "how to define success criteria" in the "ai-rag" table and displays results sorted by relevance.

Custom Embedding Function

The server implements a custom OllamaEmbeddingFunction that:

• Sends text to the Ollama API
• Receives embeddings with 768 dimensions
• Formats them for use with LanceDB

How to add this MCP server to Cursor

There are two ways to add an MCP server to Cursor. The most common way is to add the server globally in the ~/.cursor/mcp.json file so that it is available in all of your projects.

If you only need the server in a single project, you can add it to the project instead by creating or adding it to the .cursor/mcp.json file.

Adding an MCP server to Cursor globally

To add a global MCP server go to Cursor Settings > MCP and click "Add new global MCP server".

When you click that button the ~/.cursor/mcp.json file will be opened and you can add your server like this:

{
    "mcpServers": {
        "cursor-rules-mcp": {
            "command": "npx",
            "args": [
                "-y",
                "cursor-rules-mcp"
            ]
        }
    }
}

Adding an MCP server to a project

To add an MCP server to a project you can create a new .cursor/mcp.json file or add it to the existing one. This will look exactly the same as the global MCP server example above.

How to use the MCP server

Once the server is installed, you might need to head back to Settings > MCP and click the refresh button.

The Cursor agent will then be able to see the available tools the added MCP server has available and will call them when it needs to.

You can also explictly ask the agent to use the tool by mentioning the tool name and describing what the function does.