How to export Twitter (X) bookmarks

No. Bookmarks live at x.com/i/bookmarks as a list you can scroll. Settings has no “download bookmarks” control. The X bookmarks guide covers what that list is: private, unsearchable as a library, and empty wherever the original post has been removed.

So “export” always means a tool outside that screen. The three that people actually use are the account data download, a browser extension that reads the list, and saving a copy of each post as you go. They do not produce the same file.

Your bookmarks are private to your account, which is also why no outside tool can read them without your sign-in. See are X bookmarks public?

Settings and privacy → Your account → Download an archive of your data. Confirm it is you, request the archive, and wait for X to email that it is ready. The file is your account: posts, likes, messages, media and the lists you created, as X chooses to include them in that snapshot.

Do not expect a readable bookmark library. X’s description of the archive does not present it as an export of bookmarked posts, and people who open the file do not find one consistent thing. Some downloads have no bookmark file. Where a bookmark file shows up, it has been described as post IDs rather than the text. An ID is not a sentence you can read, and it will not reopen a deleted post. Use the archive as a backup of your account. Use something else as a backup of what you saved to read.

Several extensions and scripts walk the bookmarks page in your own browser and write a CSV or JSON of what loaded. That can be a useful one-time dump: URL, author, text, date, if the tool captured them. Read the permissions before you install one. It is looking at a private list.

Two limits show up in practice, and neither is printed on the box. The list has to actually load — a very long bookmark list often stops revealing older items as you scroll, so an exporter can only save what X was willing to show. And a CSV of URLs dies the same way a bookmark dies. Unless the file contains the text, you have exported addresses, not posts.

WHT.day does not log into X and pull your bookmark list down in the background. The extension saves the post you have open: text, author, images. Each save is filed on that day in your archive, which you can search later even if the post is deleted.

If you already exported a list of URLs, you can paste them into Import (up to 500 links at a time). WHT.day fetches the ones that are still public. It cannot reconstruct a post X has removed, because there is nothing left at that URL to fetch. Import is a backfill of living links, not a recovery tool.

Pick the file by the question you will ask later. A CSV is good when you want to sort, filter and count: how many saves, which authors, which month. It is a bad place to read. Columns wrap, images are URLs, and a deleted post becomes a broken link unless the text was copied into a cell.

A full-text archive is good when you want to read and find. You search a phrase and get the post, next to whatever else you saved that day. You give up the spreadsheet. Plenty of people want both: a CSV as an index, and a copy of the words for the posts that mattered. The CSV alone is not the second thing.

Any export is a snapshot. Bookmarks you add tomorrow are not in yesterday’s CSV, and posts deleted tomorrow are not recoverable from a file that only stored the URL. Repeating the export on a schedule works until the tool breaks or the list stops loading. It is also easy to forget.

Saving at the moment you decide a post is worth keeping avoids the batch job. The copy exists before anyone deletes the original. That is a habit, not a sync. If what you wanted from an export was “I will still have this in a year”, the habit is the more reliable of the two.