Skip to content
Get packs

Guide · Open source

How to use ArchiveBox: install, add URLs and browse your archive

Updated · written with AI from the project's own documentation and checked against our catalogue data

ArchiveBox is a self-hosted app that takes URLs, bookmarks, browser history or feeds from services like Pocket and Pinboard, and saves copies of each page in several formats. This guide walks through choosing an install method, starting ArchiveBox, adding links and finding your saved files. By the end you will have a working instance and know where your archive is stored.

ArchiveBoxArchiveBox/ArchiveBox · ★ 28.9k · MIT

Quick start

  1. 1Pick an install method. Choose Docker Compose if you have Docker, or uv with Python 3.13 if you prefer a terminal install.
  2. 2Create a data folder. Make a directory such as ~/archivebox/data to hold your collection.
  3. 3Start ArchiveBox. Run the Docker Compose commands or the uv install and archivebox init steps from the README.
  4. 4Finish setup in the browser. Open the admin UI at admin.archivebox.localhost:5797 to complete setup.
  5. 5Add your first URL. Run the add command with a URL, or use the browser extension to save pages.
  6. 6Browse the results. Open the public or admin UI to view snapshots, or open the files in your data folder directly.

Check what you need before installing

ArchiveBox must be installed and run on a machine you control. The README lists two routes: Docker Compose, which needs Docker, and the uv install option, which needs Python 3.13 with uv.

The README also notes that ArchiveBox uses standard tools such as Chrome, wget and yt-dlp, so the install option you choose determines how those are set up for you.

  • Docker and Docker Compose for the Docker option
  • Python 3.13 and uv for the uv option
  • A terminal you are comfortable using

Install ArchiveBox with Docker Compose

The Docker Compose option is the one the README marks as recommended. It creates a data folder, downloads a compose file, pulls the image and starts the service. New collections are initialized automatically.

After the containers start, open the admin URL in your browser to finish setup. The README also shows that you can run one-off commands through the compose service.

  • Public UI: http://web.archivebox.localhost:5797
  • Admin UI: http://admin.archivebox.localhost:5797
mkdir -p ~/archivebox/data && cd ~/archivebox curl -fsSL 'https://docker-compose.archivebox.io' > docker-compose.yml docker compose pull docker compose up -d --wait # open http://admin.archivebox.localhost:5797 to finish setup # docker compose run --rm archivebox add 'https://example.com' # docker compose run --rm archivebox help

Install ArchiveBox with uv (command line)

If you prefer a command-line install without Docker, the README gives a uv command pinned to the 0.9 release line. After installing, you create a data folder, run init and then install to set up the dependencies.

The README also offers a shortcut script that runs the same uv install for you.

uv tool install --python 3.13 --prerelease explicit --upgrade 'archivebox>=0.9.0rc0,<0.10' mkdir -p ~/archivebox/data && cd ~/archivebox/data archivebox init archivebox install # archivebox add 'https://example.com' # archivebox help # archivebox server 0.0.0.0:5797

Add URLs and imports to your archive

You can add links one at a time with the add command. The README also says you can schedule regular imports from bookmarks or browser history, social media feeds or RSS, and services like Pocket and Pinboard. A browser extension is available for saving pages as you browse.

For each URL you feed in, ArchiveBox saves a snapshot in several formats. The README lists examples such as original HTML with CSS and JS, a single-file HTML, a screenshot PNG, PDF, WARC, title, article text and favicon. Media sites like YouTube are handled differently and can yield MP3 or MP4 files, subtitles and thumbnails.

  • Feed one URL with the add command
  • Schedule imports from bookmarks, history, RSS or Pocket/Pinboard
  • Use the browser extension to save pages while you browse
archivebox add 'https://example.com'

Browse and export your archive

You can view the archive through the self-hosted web interface, the CLI, the Python API or directly from the filesystem. The README says all these methods offer matching features such as adding tags, scheduling crawls and viewing logs.

Because the data is stored in ordinary files and folders, you can read the saved copies without ArchiveBox running. The README states that the formats, including HTML, PNG, PDF, TXT, JSON, WARC and SQLite, are meant to be readable for decades.

  • Web UI for browsing and publishing your archive
  • CLI and REST API for scripts and integrations
  • Plain files and folders for long-term access

What people use it for

  • Keeping a personal backup of bookmarks or saved links
  • Preserving research papers and web sources
  • Saving evidence or pages that may change, using the archive as a record
  • Backing up media from YouTube or SoundCloud as MP3 or MP4 files, per the README
  • Keeping a git clone and README of a GitHub or GitLab project

Frequently asked questions

Is ArchiveBox free?

ArchiveBox is open source under the MIT licence, according to the repository. The README does not describe any paid hosted service, so you run it yourself.

What do I need to run ArchiveBox?

The README lists Docker for the Docker Compose option, and Python 3.13 with uv for the uv install option. You also need a machine or server where you can run these tools.

How do I install ArchiveBox?

Use the Docker Compose commands or the uv install command from the README, then open the admin UI at admin.archivebox.localhost:5797 to finish setup. The uv route also requires running archivebox init and archivebox install.

Can ArchiveBox save pages behind logins or paywalls?

The README does not establish that ArchiveBox archives pages behind logins or paywalls without extra configuration. Check the project documentation before relying on it for those pages.

How does ArchiveBox compare to Wallabag, Linkwarden, Karakeep or SingleFile?

The project lists Wallabag, Linkwarden, Karakeep and SingleFile as comparable tools. The README does not compare features, so test each against your needs. ArchiveBox focuses on saving multiple formats to plain files on your own machine or server.

Can I read my archive if ArchiveBox stops working?

Yes. The README says data is stored in ordinary files and folders and can be read without running ArchiveBox, and that formats such as HTML, PDF and WARC are meant to stay readable for decades.

New agent skills, repos and AI tools — every week

The best new skills and open-source projects, the fastest-growing repos, and one free prompt. Five minutes to read. Unsubscribe anytime.

Our verdict

ArchiveBox suits people who want to keep their own copies of saved links and are comfortable installing and running it with Docker or Python from a terminal. If you want a simple hosted bookmark service with no setup, pick a hosted tool instead. Check the documentation before relying on it for pages behind logins or paywalls.

Install options & downloads