How to Export Your X (Twitter) Bookmarks

X does not let you export bookmarks. The official data archive (Settings → Your account → Download an archive of your data) includes your posts, likes and DMs, but bookmarks are not reliably part of it — and even when data is included, an archive is a wall of JSON rather than something you can read.

So the practical options are these three.

1. Copy them out by hand

Open your bookmarks, open each post, copy the link into a spreadsheet or note. Free, works today, and fine for twenty posts.

The catch is that you end up with a list of URLs, which is the weakest possible form of an archive: if the author deletes the post, locks their account or leaves, your row becomes a dead link with no way to recover what it said.

2. The X API

There is a bookmarks endpoint, and if you are comfortable writing a script it will pull your saved posts programmatically. Realistically this means a developer account, an app, OAuth setup, and paid API access at anything beyond trivial volume. It is the right answer if you are technical and want a repeatable job; it is a lot of work for a one-off.

3. Save them into a tool that exports

The pragmatic middle ground: stop relying on X to give your data back, and save posts into something that will. This is what SocialClipper does — click the extension on a post and it captures the whole thing (text, images, author and the post it quotes), then exports to:

• JSON — full structured data, for scripting or import elsewhere
• CSV — opens in Excel, Sheets or Numbers
• Markdown — drops straight into Obsidian or Notion, with the post text quoted and images embedded

The honest limitation: there is no bulk import of your existing X bookmarks, because X does not expose them. Anything already in your bookmarks list has to be clipped once. What this fixes is everything from today onwards.

The extension is free and unlimited on your own device, with no account needed and no time limit.

What the official X archive actually contains

Worth settling this, because it is the first thing people try. Settings → Your account → Download an archive of your data gives you a zip containing your own posts, likes, direct messages, followers and ad-interest data. It arrives as JavaScript files wrapping JSON, with a bundled HTML viewer of varying usefulness.

What it is not is a tidy list of your bookmarks. Bookmarks have never been a dependable part of the archive, and even where saved data is present, the format is identifiers rather than readable posts — you would be looking at post IDs, not the text of the thing you saved. Plan for hours, not minutes, and do not count on bookmarks being in there at all.

What each export format gives you

If you clip posts into a library instead, getting them out again is a button. The three formats are deliberately different in shape:

JSON keeps everything — the full post object, author details, media references, timestamps, your labels and notes. This is the one to pick if you intend to script against it or move it somewhere else later.

CSV is the flat version for a spreadsheet: one row per post, with columns for ID, platform, URL, author name, author handle, the post text, your labels and the date you saved it. It opens directly in Excel, Sheets or Numbers, which makes it the right choice for sorting, filtering or handing to someone who does not want a tool.

Markdown is written for note apps rather than for reimport. You get YAML frontmatter that Obsidian and Notion both read, a #tag line per entry, each post blockquoted under its author's name with a link to the original, and a horizontal rule between entries so splitting one file into per-note files is trivial.

You can export a single label rather than the whole library, which is how most people use it: export research as Markdown, drop it into Obsidian, and leave the rest where it is.

Why this matters more than it sounds

An export is the test of whether you own your library. A tool that cannot hand your data back has not stored it for you — it has borrowed it. X's bookmarks fail that test, which is the real reason the question “how do I export my X bookmarks” gets asked so often and answered so badly.

The second-order point is that an exportable library is portable. If something better comes along, or you want your saves in a personal wiki, or you simply want a file in a folder that outlives every product involved, that is a download rather than a project.

A realistic plan for the backlog

Three steps that work:

• Check what still exists. A long bookmark list has already lost entries to deletions and suspensions. Scroll it once and notice how many rows say the post is unavailable — that tells you what the backlog is really worth.
• Clip the ones you have genuinely returned to. Usually a dozen or two. Those are the ones with real value, and clipping captures the content so they stop being at risk.
• Change where new saves land. This is the part that compounds. Everything from today onwards is in a library you can search and export.

If you want more on the retrieval side rather than the export side, organising your X bookmarks covers labelling, and searching them covers what X's own search can and cannot do.