Blog › Bookmark media download

Download Every Photo & Video From Your X Bookmarks (Original Quality)

Most "export bookmarks" tools hand you a spreadsheet of URLs. That feels like a backup right up until the day a post gets deleted — and every media link in your spreadsheet turns into a 404. A real media backup means the actual image and video files, saved to your own disk, at the quality they were uploaded. Here's why that matters and how to do it.

Key points

  • URLs ≠ files. An exported link is a pointer to X's CDN. Delete the post (or the account) and the pointer dies.
  • Link rot is the norm on X. Viral art threads, reference dumps and tutorials get deleted constantly — often within months.
  • Default images aren't full size. pbs.twimg.com URLs need ?name=orig for the original resolution; videos have multiple quality variants and you want the highest-bitrate MP4.
  • The practical route: a Chrome extension that saves the actual media bytes from your bookmarked posts to your Downloads folder — no servers, no API keys.

Why a CSV of links fails as a media backup

Export your bookmarks to CSV and you'll get columns like author, text, date, and maybe a photo_url. Open it a year later and click a media link: a meaningful share of them are dead. The causes stack up:

A backup that depends on someone else's server staying up, and someone else's account staying public, isn't a backup. It's a wish list.

The quality trap: what X actually serves you

Even when a link works, what you get may not be what was uploaded.

Photos: the name=orig parameter

X serves images through pbs.twimg.com with size variants. A bare URL often lands on a downscaled version (small, medium, large). Appending ?name=orig requests the original upload:

https://pbs.twimg.com/media/XXXXX.jpg?name=orig

Without it, a 4000×3000 photo can arrive as a ~1200px JPEG — fine on screen, painful if you ever wanted to print, crop, or re-examine detail (reference images, charts, screenshots of code).

Videos and GIFs: variants, not a single file

Video on X is adaptive: each video has multiple MP4 renditions at different resolutions/bitrates, plus an HLS playlist. The DOM usually plays a mid-tier variant. A naive "save video" grabs whatever buffer is playing — not the 1080p file. GIFs aren't GIFs at all; they're silent looping MP4s. Doing this properly means picking the highest-bitrate MP4 variant per video.

Route 1: The X API (paid, capped, no files)

The developer API's bookmarks endpoint returns media metadata — photo URLs and video variant lists. In theory you could script a downloader. In practice:

It's a weekend project for an engineer and a dead end for everyone else.

Route 2: Right-click / third-party downloaders (manual, lossy)

You can save images one by one from the browser, or use generic video-download extensions on each bookmarked post. This works for a dozen posts; it doesn't scale to hundreds, and generic downloaders typically grab the currently-playing video variant rather than the best one. There's also no structure — you end up with image (37).jpeg in Downloads with no link back to which post it came from.

Route 3: A bookmarks-aware archiver (what we built)

The practical middle path is an extension that understands bookmarks as a collection: it watches your bookmarks page load, records each post with its author/date/text, resolves media to the best available variant, and writes the real files to your disk with a structured export alongside.

What a good implementation gets right:

Archive bookmarked media before it's deleted

XVault downloads the actual photos and videos from your X bookmarks in original quality, plus full CSV/JSON/Markdown exports — all to your own device, no servers, no API keys. Free up to 1,000 rows per data type per export; PRO ($19.99 one-time) removes the cap.

Get XVault for Chrome

Organizing the archive so it stays useful

A pile of IMG_0001.mp4 rots almost as badly as links. Two habits make a media archive actually durable:

  1. Keep the index with the files. Store the CSV/JSON export in the same folder tree. Post ID, author handle, date, text — that metadata is what lets you find "the thread about font licensing" two years later.
  2. Back up the backup. An archive on one disk is one disk failure away from zero. Sync the folder to a NAS or cloud drive of your choice. The files are yours; put them wherever your own backup routine already runs.

Because the media are plain JPEG/PNG/MP4 files, everything downstream is ordinary: Spotlight/Everything search, photo viewers, video players, Obsidian embeds. No proprietary vault, no account required to read your own archive.

Frequently asked questions

Can I download media from bookmarks in bulk without an extension?

Only via the paid API plus a custom script, and you'll still hit the ~800-bookmark pagination wall. There's no official bulk media download for bookmarks.

Why do some bookmarked videos fail to download?

The post was deleted, the account went private/was suspended, or the media is region/age-restricted. Once the source post is gone, no tool can fetch the file — which is the argument for archiving early.

Are saved videos the original upload?

X re-encodes all uploads. The best anyone can retrieve is X's highest-bitrate rendition (typically up to 1080p); the uploader's raw file stays with them. name=orig photos, however, are the full-resolution image X stores.

Is downloading bookmarked media legal?

Saving copies of publicly posted content for personal archival use is widely practiced, but copyright stays with the creator. Keep archives personal — redistribution or commercial reuse needs the creator's permission.

Does this work for likes and profiles too?

With XVault, yes — likes, your own tweets/replies, followers/following, and media across them, exported the same local-first way.