Skip to content
Live demo

Introduction

SparkFeed is a modern, self-hosted RSS reader designed for people who value privacy, speed, and simplicity. It is an open-source product built by Sparkable.

Instead of switching between multiple websites, SparkFeed brings all your favorite blogs, news, and updates into one clean interface.

It runs entirely on your machine using Bun and stores data in a local SQLite database, or a remote Turso database for production deployments.

No ads. No tracking. Just content.

SparkFeed is built for anyone who wants full control over what they read and how they read it:

  • Developers: who want a self-hosted, hackable feed reader
  • Students: who want to follow research blogs, newsletters, and communities
  • Founders: who want to stay on top of industry news without the noise
  • Content creators: who want to turn reading into a content creation workflow powered by Spark AI

SparkFeed is a full-stack web application that runs on your machine. Here’s the high-level flow:

  1. You add RSS/Atom feed URLs through the UI
  2. SparkFeed fetches and parses those feeds in the background
  3. Articles are stored locally in a SQLite database
  4. You read articles in a clean, distraction-free interface
  5. Spark AI summarizes your feeds and answers questions about your content

Every article, feed, and preference is stored on your machine. There’s no server to go down, no account to lose, and no data sent anywhere.

SparkFeed makes no outbound requests except the RSS feed URLs you explicitly add. There’s no telemetry, no analytics, and no third-party services.

The UI is designed to get out of your way. Navigation is instant thanks to TanStack Router’s client-side routing and the local SQLite database.

SparkFeed is open source. The codebase is clean, well-typed TypeScript, and designed to be easy to extend with custom features.

Ready to get started? Move on to What is SparkFeed? for a deeper look at the project, or jump straight to Installation if you’re ready to run it.