---
title: "X bookmarks in Keep, not in a private list | Keep"
description: "Why X bookmarks are a pointer, not a library, and how to keep the useful posts as searchable Markdown in Keep. Capture, import, and X Sync."
canonical: "https://keep.md/blog/twitter-bookmarks-guide"
language: "en"
---

# X bookmarks in Keep, not in a private list

Why X bookmarks are a pointer, not a library, and how to keep the useful posts as searchable Markdown in Keep. Capture, import, and X Sync.

April 22, 2026Ian Nuttall

X bookmarks feel like a library. They are a private pointer list. The post still lives on X. If the author deletes it, goes private, or the account disappears, your bookmark becomes an empty row.

This page is about treating bookmarks as source material you actually keep. Privacy, public bookmark counts, and the “did X add a hard cap” arguments live on [ilo’s bookmarks guide](https://ilo.so/blog/twitter-bookmarks-guide). Same keywords, different job.

## What you actually wanted when you hit bookmark

Usually one of three things.

1.  Find this post again without scrolling `x.com/i/bookmarks`.
2.  Use it later in writing, research, or an agent session.
3.  Still have the text if the original post dies.

The native list is good at capture. It is weak at all three follow-ups. Search only looks at tweet text. Folders (Premium) are named buckets, not a second copy. There is no jump-to-date. There is no export in Settings.

So the useful move is to copy the posts you care about into a system that stores content, not just URLs.

## Three different products for that move

**A file.** CSV or JSON on disk. Best when you want proof, a spreadsheet audit, or an import into something else. [ilo](https://ilo.so/) does this in Chrome. Details in the [export into Keep](https://keep.md/blog/export-x-bookmarks) post.

**A social bookmark app.** Dewey if your whole habit is posts across X, LinkedIn, Bluesky, TikTok, and the rest. Nested folders, auto-tags, social-shaped UI. Weak the moment you also save articles, PDFs, and videos.

**A mixed Markdown library.** Keep. X posts sit beside web articles, YouTube transcripts, feeds, files, and notes. Search hits the saved content. Agents can read the same items over [MCP](https://keep.md/docs/mcp).

Pick the product that matches the job. Do not pick Keep because a leftover article still claims the Keep extension dumps your entire X history. That work moved.

## How Keep captures X

**Public posts, any Personal plan.** Save from Capture, [Web Clipper](https://keep.md/docs/extension), CLI, API, or MCP. Keep can expand a saved post into the author’s thread.

**Private bookmark list.** The [X Sync add-on](https://keep.md/docs/x) connects your account and imports new bookmarks. You can start from now or run a capped recent backfill. It is not a full historical replay.

**Old library you already have.** Export with ilo, Bookmark Save, or Dewey, then [import the file](https://keep.md/docs/import). CSV and generic JSON use the normal bookmark importer. A compatible Keep-format JSON file can keep richer post fields.

Once an item is in Keep, it is Markdown on storage you control. X going sideways later does not wipe the copy you already saved.

## Search that is not X’s bookmark box

X bookmark search is “does this tweet body contain the string.” It does not search the linked article, the thread under the post, or a note you meant to leave yourself.

Keep search is over the saved Markdown. Tag by intent (“use in the launch doc”, “read this week”) instead of filing into X folders named after topics. Folders help on X until the list is large. They do not help you find “that onboarding thread” six months later.

If you highlight while you read, Keep’s highlight tools are newer and lighter than Readwise Reader. Reader still wins a daily-review, highlight-first habit. Keep wins when the post has to live next to the rest of your research.

## Dewey, ilo, Reader, Keep

| If you want | Use |
| --- | --- |
| CSV or JSON and stop there | ilo |
| Social posts across several networks, one dedicated UI | Dewey |
| Highlighting, review queue, threads as long-form | Readwise Reader |
| Mixed sources as Markdown, including agents | Keep |

[Dewey alternatives](https://keep.md/alternatives/dewey-alternatives) and [Dewey vs Keep](https://keep.md/compare/dewey-vs-keep) spell out the social-only vs mixed-library split. Raindrop is the polished bookmark manager if you care more about collections than Markdown.

## A working setup

1.  Export the old X list once. Keep the file.
2.  Import a sample into Keep and spot-check old, tagged, and media-heavy rows.
3.  Turn on Web Clipper or X Sync so new bookmarks do not depend on another scrape.
4.  Prune. A few hundred items you can search beats three thousand you never open.

Bookmarking is capture. Capture without a second home is a pile that X can still delete out from under you.

[Start Keep for free](https://app.keep.md/signup?utm_source=keep.md&utm_medium=content&utm_campaign=twitter-bookmarks-guide-inline) and save the next useful post into the library instead of leaving it on X.

## Keep anything useful as Markdown.

Save notes, links, pages, files, and other useful material as clean Markdown. Read it later, search it, or use it with an AI agent.

[Start for free](https://app.keep.md/signup?utm_source=keep.md&utm_medium=content&utm_campaign=blog-twitter-bookmarks-guide)

[Read the docs](https://keep.md/docs)