OpenMetadata AI Explorer

A VS Code/Cursor extension that brings intelligent data discovery directly into your IDE. Search your OpenMetadata catalog using natural language, get Gemini-powered insights, and visualize data relationships without switching contexts.

📖 Read the full development article: Building a Gemini-powered data catalog extension for Cursor - Learn how this extension was built and the technical decisions behind it.

Demo

Watch the extension in action:

What It Does

• Natural Language Search: Ask questions like "show me customer tables" or search by keywords
• Gemini AI Insights: Get intelligent analysis of your datasets and data quality
• Interactive Data Lineage: Visualize upstream and downstream table relationships
• Column Details: Explore table schemas with expandable column information
• IDE Integration: Works seamlessly next to the Cursor terminal panel

Requirements

1. OpenMetadata Server: Running locally at http://localhost:8585

   • Use the OpenMetadata Docker setup
   • Load sample data for testing

2. Gemini API Key: Optional but recommended for AI features

   • Get a free key from Google AI Studio
   • Keys start with AIza...

Installation & Setup

Option 1: Install Pre-built Extension (Recommended)

1. Download the Extension Download the latest release: openmetadata-ai-explorer-1.1.0.vsix

2. Install in Cursor

• Open Command Palette (Ctrl+Shift+P or Cmd+Shift+P)
• Type: Extensions: Install from VSIX...
• Select the downloaded .vsix file
• Reload when prompted

3. Configure Extension Settings Open Cursor settings (Ctrl+, or Cmd+,) and add:

{
  "openmetadataExplorer.openmetadataUrl": "http://localhost:8585",
  "openmetadataExplorer.geminiApiKey": "YOUR_GEMINI_API_KEY",
  "openmetadataExplorer.openmetadataAuthToken": "YOUR_BOT_TOKEN"
}

4. Get OpenMetadata Bot Token

1. Open http://localhost:8585 and login (admin/admin)
2. Go to Settings → Bots
3. Click Add Bot with these details:
   • Name: cursor-extension-bot
   • Description: Bot for Cursor extension
4. Click Generate Token and copy the JWT token (starts with eyJ)
5. Assign Data Consumer role to the bot

5. Start Using

• Look for OpenMetadata AI panel at the bottom
• Verify connection status shows green checkmarks
• Try searching for "customer" or "orders"

Option 2: Development Mode (Build from Source)

1. Clone and Install

git clone https://github.com/hugozanini/open-metadata-cursor-extension.git
cd open-metadata-cursor-extension
npm install

2. Configure Settings Follow steps 3-4 from Option 1 above to configure the extension settings.

3. Run in Debug Mode

1. Press F5 to launch the extension in a new Cursor window
2. Look for OpenMetadata AI panel at the bottom
3. Verify connection and start searching

How to Use

Search Your Data

• Keyword Search: Type table names like "customer" or "orders"
• Natural Language: Ask questions like "show me customer data"
• Browse Results: Click on tables to see column details

View Data Lineage

1. Search for any table
2. Click View Lineage on the table card
3. Use the interactive graph:
   • Click + buttons to expand upstream/downstream relationships
   • Click - buttons to collapse connections
   • Drag nodes to reposition them
   • Zoom with mouse wheel

Example Searches

• customer - Find customer-related tables
• orders - Discover transaction data
• sales - Locate revenue tables
• product - Find catalog information

Development

Build Commands

# Development build with watch
npm run watch

# Production build
npm run compile

# Package for distribution
npm run package

Project Structure

• src/extension.ts - Main extension entry point
• src/services/ - OpenMetadata API and Gemini integration
• src/webview/ - React components for the UI
• src/webview/components/Lineage/ - Data lineage visualization

Configuration Options

Setting	Description	Default
openmetadataExplorer.openmetadataUrl	OpenMetadata server URL	http://localhost:8585
openmetadataExplorer.geminiApiKey	Gemini API key for AI features	(empty)
openmetadataExplorer.openmetadataAuthToken	OpenMetadata bot JWT token	(empty)

Current Status

Version 1.1 - Production Ready

• Natural language search with Gemini AI
• Interactive data lineage visualization
• Professional UI optimized for developers
• Enterprise-ready VSIX distribution

Planned Features

• Column-level lineage relationships
• Data quality monitoring integration
• Advanced search filters and exports

Contributing

Open a Pull Request!

License

MIT License