Video coming soon…

🐿️ Setup Hoarder — AI-Powered Bookmark Manager

Deploy a self-hosted bookmark manager with automatic AI tagging, full-text search, and screenshot archiving for links and notes.

⚠️ This script is provided for demo and testing purposes only. Not intended for production use.

📦 Resources & Setup Scripts

Grab the automated bash script from GitHub to follow along with the video.

Automated install — Hoarder bookmark manager running in one command.
View on GitHub

Quick Install:

wget https://raw.githubusercontent.com/mhmdali94/Docker/main/tools/hoarder/hoarder-ubuntu.sh
chmod +x hoarder-ubuntu.sh
sudo bash hoarder-ubuntu.sh

Tutorial Steps

1 Download & Run the Installer

The script installs Docker, pulls the Hoarder, Meilisearch, and Chrome headless images, generates a secure secret key, and starts all containers. Hoarder will be available on port 3000 with Meilisearch providing fast full-text search on port 7700 (internal only).

wget https://raw.githubusercontent.com/mhmdali94/Docker/main/tools/hoarder/hoarder-ubuntu.sh
chmod +x hoarder-ubuntu.sh
sudo bash hoarder-ubuntu.sh

2 Access the Web UI

Open your browser and navigate to Hoarder:

http://<your-server-ip>:3000

3 Add Your First Bookmarks

Register your account and start saving resources. Paste any URL into the add bar to bookmark a web page — Hoarder will fetch the page title, description, and take a screenshot automatically. You can also save plain text notes and images. Import your existing bookmarks from a browser HTML export via Settings → Import.

4 Configure AI Tagging

Go to Settings → AI and add your OpenAI API key to enable automatic tag generation. Once configured, every saved bookmark will be analyzed by AI and tagged with relevant keywords automatically. Alternatively, set up Ollama for free local AI tagging using open-source models — no API key required.

Ports Used

PortPurpose
3000Hoarder Web UI
7700Meilisearch (internal)

Overview

Hoarder is a self-hosted bookmark manager with AI-powered automatic tagging using OpenAI or Ollama. It saves full page snapshots, extracts readable content, and provides full-text search across all saved bookmarks. It also supports browser extensions and a mobile app for quick saving.

Why Use It

Unlike browser bookmarks or cloud services like Pocket or Raindrop.io, Hoarder runs entirely on your server. Your reading list and saved articles are private, searchable, and never lost even if a site goes offline — because Hoarder archives the full page content.

When You Need It

    Who Should Use It

      Real Use Cases

        Main Features

          How to Use After Installation

            Security Best Practices

              Ports and Firewall Notes

              Hoarder runs on port 3000 by default. Place it behind a reverse proxy with SSL for public access. If used personally, keep it on a local network and access via VPN to avoid any public internet exposure.

              Backup and Maintenance

                Common Mistakes

                  Troubleshooting

                    Alternatives

                    Alternatives include Pocket (cloud, limited free tier), Raindrop.io (cloud, polished UI), Wallabag (self-hosted, simpler), and Linkding (self-hosted, lightweight). Choose Hoarder when you want AI auto-tagging plus full-page archiving in a self-hosted setup.

                    When Not to Use It

                    Avoid Hoarder if you need a minimal, resource-light bookmark manager — Linkding or Shiori would use far less RAM. Also avoid if you are not comfortable managing a multi-container stack that includes Hoarder, Meilisearch, and optionally Ollama.

                    PrismaTechWork Professional Help

                    PrismaTechWork provides end-to-end infrastructure services — from initial deployment and security hardening to ongoing monitoring, automated backups, and dedicated support. Whether you need a single-server setup or a multi-site network, our team ensures your infrastructure is built right, secured properly, and maintained reliably.

                      Contact Us

                      Frequently Asked Questions

                      Does Hoarder require an OpenAI API key?

                      No — AI tagging is optional. You can run Hoarder without an AI provider and tag bookmarks manually. If you want auto-tagging, you can use an OpenAI API key (paid) or point Hoarder to a local Ollama instance (free, runs locally).

                      How does Hoarder archive pages?

                      When you save a bookmark, Hoarder fetches the page content, extracts readable text similar to a reader mode, and takes a screenshot. This archived content is stored in your database and storage volume, so the bookmark remains searchable even if the original URL is deleted.

                      Is there a browser extension for Hoarder?

                      Yes. Hoarder has official browser extensions for Chrome and Firefox. Install the extension, point it to your Hoarder instance URL, and authenticate. You can then save any page with a single click from the browser toolbar.

                      Can I import bookmarks from Pocket or my browser?

                      Yes. Hoarder supports importing bookmarks from Pocket JSON exports and browser bookmark HTML exports. Use the Import feature in the settings page to bulk-import your existing collection.

                      What is Meilisearch and why does Hoarder need it?

                      Meilisearch is a fast open-source search engine that Hoarder uses to index and search bookmark content. It enables instant full-text search across all your saved pages. The Docker Compose setup starts Meilisearch automatically alongside Hoarder.

                      Can I access Hoarder from my phone?

                      Yes. Hoarder has mobile apps for iOS and Android. The mobile app lets you share URLs directly to Hoarder from any app on your phone via the system share sheet. You can also access the web UI from your phone browser.

                      How do I use Ollama for AI tagging instead of OpenAI?

                      Install Ollama on the same server or a nearby machine and pull a model such as llama3. Set the OLLAMA_BASE_URL environment variable to your Ollama server URL and configure OLLAMA_MODEL in Hoarder's environment. Hoarder will then use your local LLM for tagging at no cost.

                      Can multiple users use the same Hoarder instance?

                      Hoarder supports multiple user accounts. Each user has their own private bookmark collection. The admin creates additional accounts from the admin panel. There is no public self-registration by default — the admin must invite users.