AppKit
Build Databricks Apps faster with a Node.js + React SDK. Built for humans and AI.
AppKit is a TypeScript SDK for building production-ready Databricks applications with a plugin-based architecture. It gives you opinionated defaults, built-in observability, and first-class integration with Databricks services — so you spend time on your app, not the plumbing.
- Plugin architecture — Compose your app from focused plugins; a base server plus analytics, Genie, files, and more.
- Type safety — End-to-end TypeScript with automatic query type generation.
- Production-ready — Built-in caching, telemetry, retries, and error handling.
- Great DX — Hot reload, file-based queries, and a workflow tuned for AI-assisted development.
- Databricks native — SQL Warehouses, Unity Catalog, Lakebase, and other workspace resources out of the box.
Quickstart
Requires the Databricks CLI and Node.js.
AI-assisted (recommended) — scaffold and evolve your app by prompting an AI assistant. Install the Agent Skills once:
databricks aitools install
Then ask your assistant to build the app, e.g. "Create a Databricks app that shows a dashboard of the NYC taxi trips dataset." See the AI-assisted development guide.
Manual — scaffold and deploy with the CLI directly:
databricks apps init # scaffold the app + install dependencies (follow the prompts)
databricks apps deploy # deploy to your workspace
What an app looks like
Every app is a set of plugins hosted by server():
import { createApp, server, analytics } from '@databricks/appkit';
createApp({
plugins: [server(), analytics()],
});
With analytics(), drop a .sql file in config/queries/ and it's served at POST /api/analytics/query/<query_key> — parameterized, cached, and (with a .obo.sql name) executed on behalf of the signed-in user.
Plugins
AppKit's power comes from its plugin system. Each plugin adds a focused capability — analytics, genie, files, and more — with minimal configuration. See the plugins documentation for the full list and API.
Missing a plugin? Open an issue and tell us what you need — community input directly shapes the roadmap.
Documentation
- AppKit documentation — getting started, guides, and full API reference.
- AI-assisted development — build and evolve apps with an AI assistant.
- New to the Databricks Apps platform? Start with the Apps quickstart.
Contributing
See CONTRIBUTING.md for development setup and contribution guidelines.