Self-Hosted Web Archiving & Page Capture

The web rots. Articles get paywalled, repos get deleted, tweets get retconned. This category is for capturing pages on your own hardware so future-you can still read them. For bookmark / read-later managers see Self-Hosted Bookmarks & RSS; for personal media servers see Self-Hosted Photos & Media.

Full archive servers

• ★ ArchiveBox — the default. Takes URLs (or RSS / Pocket / Pinboard / browser exports) and saves WARC + screenshot + PDF + readability HTML + media + git source — multiple formats per URL. Mature; web UI; CLI.
  • https://github.com/ArchiveBox/ArchiveBox
• ★ Karakeep (formerly Hoarder) — primarily a bookmark manager but archives full pages with screenshots and runs OCR + AI tagging. Great if you want one tool for "save link + archive + read later." See also Self-Hosted Bookmarks & RSS.
  • https://github.com/karakeep-app/karakeep
• LinkAce — bookmark manager with Wayback-Machine-link-on-save; lightweight archive.
  • https://github.com/Kovah/LinkAce
• Linkwarden — bookmarks + page archive (PDF + screenshot + readable); Karakeep alternative.
  • https://github.com/linkwarden/linkwarden

Single-page capture (one URL → one file)

• ★ SingleFile — browser extension + CLI; saves a complete page (CSS, images, fonts) into a single self-contained .html file. The cleanest "freeze this page" output.
  • https://github.com/gildas-lormeau/SingleFile
• SingleFileZ — same author; ZIP-format alternative, smaller files.
• monolith — Rust CLI, "save complete web pages as a single HTML file." Pairs well with cron / scripts.
  • https://github.com/Y2Z/monolith
• wget --mirror / httrack — old-school site mirroring; still useful for whole-site captures.
• curl-impersonate — for sites that block plain curl; impersonates real browser TLS fingerprints.

WARC-format archiving

• ★ wget --warc-file=... — produces standard WARC archives readable by Wayback / OpenWayback / pywb.
• Browsertrix Crawler (Webrecorder) — Playwright-based crawler that produces WARC; handles JS-heavy modern sites; captures network responses faithfully.
  • https://github.com/webrecorder/browsertrix-crawler
• Browsertrix Cloud — multi-user crawl-management UI on top.
  • https://github.com/webrecorder/browsertrix
• ReplayWeb.page — view WARC files in a browser; serves your archive without a server.
  • https://github.com/webrecorder/replayweb.page
• pywb (Python Wayback) — host your WARCs as a private Wayback Machine.
  • https://github.com/webrecorder/pywb

Page change monitoring

• ★ changedetection.io — watches URLs for changes; CSS / xpath / JSON filters; notifications via Apprise; great for "alert me when this product is back in stock" or "alert me when this docs page changes." The default in 2026.
  • https://github.com/dgtlmoon/changedetection.io
• urlwatch — CLI; YAML config; venerable.
  • https://github.com/thp/urlwatch
• diff2html-cli + cron + git — DIY pattern: clone the page text, commit changes, get a diff in git history.
• Distill.io / Visualping — closed-source SaaS; widely used.

Media downloaders (bulk capture)

• ★ gallery-dl — image / gallery downloader for hundreds of sites (Twitter, Reddit, Instagram, DeviantArt, etc.). The "save my likes" tool.
  • https://github.com/mikf/gallery-dl
• ★ yt-dlp — video downloader; backbone of many self-hosted archives.
  • https://github.com/yt-dlp/yt-dlp
• MeTube / Pinchflat — web UIs around yt-dlp for "subscribe to a channel, mirror new videos."
  • https://github.com/alexta69/metube
  • https://github.com/kieraneglin/pinchflat
• TubeArchivist — full YouTube archive with metadata + thumbnails; Plex / Jellyfin-friendly.
  • https://github.com/tubearchivist/tubearchivist
• wallabag for read-later text articles — see bookmarks page.

Wayback Machine integration

• waybackpack / wayback-machine-downloader — pull a domain's history from the Internet Archive's Wayback for local storage.
• Save Page Now API — push your URLs to the public Wayback Machine; many bookmark tools call this on save.

Patterns to adopt

• ★ Auto-archive on bookmark. Karakeep, ArchiveBox, and Linkwarden all support "every saved URL is also archived."
• Submit to public Wayback too — redundancy via Internet Archive on save.
• Object storage for archives — WARCs and screenshots are large. Back archive volumes to S3 / R2 with restic.
• Don't archive logged-in pages by default — cookies might leak. Configure profile capture deliberately.
• Cron RSS-to-archive — pipe a curated RSS feed straight into ArchiveBox to mirror everything you read.
• Consider law / ToS — broad mirroring of paywalled or private content has legal limits; archive what you have rights to.

Pick this if…

• Default full archiver, many formats: ArchiveBox.
• Bookmarks + archive in one tool: Karakeep or Linkwarden.
• One-off "save this page": SingleFile (browser) or monolith (CLI).
• JS-heavy modern sites / WARC fidelity: Browsertrix Crawler + ReplayWeb.page.
• Watch a URL for changes: changedetection.io.
• Bulk media (videos / images): yt-dlp + gallery-dl, with MeTube / TubeArchivist as front-ends.
• Private Wayback Machine: pywb on top of WARCs from Browsertrix.