hoarder-app/hoarder: A self-hostable bookmark-everything app (links, notes and images) with AI-based automatic tagging and full text search

A self-hosted bookmarking app with artificial intelligence capabilities for data hoarders.

• 🔗 Bookmark links, take simple notes and save images and pdfs.
• ⬇️ Automatically obtain link titles, descriptions and images.
• 📋 Sort bookmarks into lists.
• 🔎 Full-text search of all stored content.
• ✨ Automatic tagging based on artificial intelligence (also known as chatgpt). Support local models using ollama!
• 🎆 OCR is used to extract text from images.
• 🔖 Chrome plug-in and Firefox plug-in Used to quickly add bookmarks.
• 📱 Ann iOS Appand a Android App.
• 📰 Automatically accumulate RSS feeds.
• 🌐 REST API.
• 🗄️Full page archive (use boulder) to prevent link failure. Automatic video archiving usage youtube-dl.
• ☑️Batch operation support.
• 🔐Single login support.
• 🌙 Dark mode support.
• 💾 Self-host first.
• [Planned] Download content for offline reading.

⚠️ This app is under heavy development and is still unstable.

You can access the demo: https://try.hoarder.app. Log in using the following credentials:

email: demo@hoarder.app
password: demodemo

The demo contains some content, but it is in read-only mode to prevent abuse.

• NextJS For web applications. Use application router.
• drizzle Used for databases and their migrations.
• Next verification Used for authentication.
• remote procedure call Used for client->server communication.
• puppet master Used to grab bookmarks.
• Open artificial intelligence Because artificial intelligence is so popular right now.
• beauty search For complete content search.

I often browse reddit, twitter, and hackernews on my phone. I often find interesting things (articles, tools, etc.) that I want to bookmark and read later on my laptop. Typical read later application use case. Initially, I was using pocket to this end. Then I got into self-hosting and I wanted to self-host this use case. I have used memorandum I like it a lot for those quick notes, but it’s missing some features that I think are important for this use case, such as link preview and auto-tagging (more on that in the next section).

My day job is as a systems engineer (and has been for the past 7 years). I don’t want to get too removed from the web development world. I decided to build this app as a way to involve myself in web development while building something I care about and use every day.

• memorandum: I like notepads. I have it running on my home server and it is one of my most commonly used self-hosted applications. However, it does not archive or preview the links shared within it. It’s just that I dump a lot of links in there and I’d be happy if I could figure out which link is that by looking at my timeline. Also, considering the variety of stuff I dump there, I’d love it if it did some sort of automatic tagging of the stuff I save there. This is exactly the use case I’m trying to solve with Hoarder.
• my heart:Mymind is the closest alternative to this project and I draw a lot of inspiration from it. Although it is a commercial product.
• raindrop: A beautiful open source bookmark manager that supports links, images, and files. But it’s not self-hosted.
• Bookmark Manager (mainly focused on bookmark links):
  • pocket: Pocket got me interested in Read Later apps. I have used a lot of. However, I recently got into home labs and became obsessed with the idea of ​​running my services in my home server. Hoarder aims to be the first self-hosted application.
  • Link Warden: An open source self-shelf bookmark manager that I ran for a while in my home lab. It focuses primarily on links and supports collaborative collections.
  • omnivore: Omnivore is a really cool open source read later app. Unfortunately, it relies heavily on Google Cloud infrastructure, which makes hosting your own website very difficult. They published a Blog article About how to run a minimal omnivore, but it’s missing a lot of things. Self-hosting didn’t seem to be a high priority for them, and it was something I cared about, so I decided to build an alternative.
  • wallabagh: Wallabag is a full-fledged open source read-later application written in php, which I think is a common recommendation for this type of application on reddit. To be honest, I haven’t really tried it and the UI feels a bit dated. Honestly, it’s probably more stable and full-featured than this app, but where’s the fun in that?
  • motto:Shiori is an open source pocket clone written in Go. It ticks all the boxes except for my super complex AI based markup. (JK, I only discovered this after I decided to build my own app, so here we are 🤷).

Hoarder uses Weblate to manage translations. If you want to help translate Hoarder you can do this here.

If you like using Hoarder, please give the repository a ⭐️!