How to search your X (Twitter) likes

Import the posts you want to search

Likes Library searches the text stored in your account. It does not search all of X or read another person’s private likes. Start by creating a library and connecting the Chrome or Firefox extension. Choose Import my likes and keep the X import tab visible while it collects the posts X makes available.

If you already have an archive, upload the relevant archive files instead. You can use both methods: a post with the same X ID is merged rather than added twice. Once posts appear in your dashboard, you can search them while the rest of an import continues.

Start with a distinctive word, then narrow the results

Use the search field above the saved posts. A tool name, an unusual noun, or a short piece of wording is usually a better starting point than a broad subject such as work. For example, try onboarding, then add checklist if you remember both words appearing in the post.

Every word you enter must occur in the saved text. Words can appear in any order, and matching ignores letter case and accents. Adding a word narrows the results; removing one broadens them. Search matches parts of words too, so a short fragment can help when you do not remember the ending.

This is text matching, not a question-answering system. Do not add quotation marks, OR, or X search operators such as from:; Likes Library does not interpret them as special commands. Try alternative words in separate searches.

Search the whole library or one author

Choose All liked posts when you want results across the collection. If you remember the writer, open them in Authors and choose This author. This is useful for a phrase a person has used more than once, or for reading their saved posts around a recurring subject.

The Authors view ranks writers by the number of their posts you have saved. That can surface a familiar person even when you have forgotten their handle. If a search looks unexpectedly empty, check its scope before changing the words.

Use topics when you remember the idea

With optional AI categories enabled, Topics groups related saved posts. A topic can bring together posts that discuss the same idea with different wording. Open a relevant group, read the posts, and use a distinctive term you find there for your next search.

Topics and keyword search serve different purposes. Search requires matching words; a category is an AI interpretation of the text. Categories can be mistaken and corrected. Search remains available with AI turned off. See how AI categories work before enabling them.

Check the saved text before assuming a post is gone

A remembered sentence might have been in a linked article, an image, a reply, or another post in a thread. Those words may not be in the particular post you liked. Likes Library does not make text inside images or the contents of every linked page searchable.

Try another clue and check whether the post was imported. Shortened source text can also limit matching; the extension tries to obtain fuller text where X makes it available. For a longer investigation, follow the old liked tweet workflow or the missing likes checklist. You can practice with fictional posts in the interactive demo.