Back to all guides
Bookmarks
October 22, 2026
7 min read

How to Find Dead Bookmarks in Chrome

Chrome has no idea whether your bookmarks still work, and it will never volunteer the information. Here is how to find out.

Chrome stores a bookmark as a title and a URL. It never checks either again. A bookmark saved in 2018 to a site that shut down in 2021 looks exactly like one you made this morning, and you find out which is which by clicking.

There are three ways to audit them. They differ in effort and in what they catch, and the differences matter more than you would think.

Option 1: by hand

Open the Bookmark Manager with Ctrl+Shift+O (Cmd+Option+B on a Mac), then middle-click your way through, closing tabs as you go.

This is the only method that catches everything, because you are the checker. You will notice the parked domain, the paywall, the page that loads but has been replaced with something unrelated. You will also give up around bookmark ninety, which is the problem — most libraries worth auditing are far bigger than anyone will click through.

Option 2: a script over your export

Export your bookmarks to HTML, then have something walk the list and request each URL. If you are comfortable in a terminal this is twenty minutes of work and it scales to thousands of links.

What it catches: 404s, dead domains, connection failures. What it misses is the interesting part.

Why status codes lie. A parked domain returns 200. A paywall returns 200. A sign-in wall returns 200. A site that replaced your article with a generic category page returns 200. Checking status codes alone will clear a large share of bookmarks that are, in every sense you care about, dead.

The other practical problem is that sites increasingly refuse automated requests. A script hitting a few hundred URLs from one address starts collecting 403s and rate limits, which look identical to genuine failures unless you go through them individually.

Option 3: a bookmark checker

A dedicated checker does what the script does, but handles the awkward parts: spacing requests so you are not mistaken for a scraper, following redirects to see where they actually land, and distinguishing “gone” from “refused to answer a robot”.

Ours is free and runs in your browser. That last part is not a marketing line — your bookmark file is a detailed record of what you read and care about, and uploading it to a stranger’s server to get a list of broken links back is a poor trade. Nothing leaves the page.

Reading the results honestly

Whatever you use, three categories need judgement rather than automation:

  • Redirected — the link works but now lands somewhere else. Sometimes that is a tidy site reorganisation; sometimes it is a domain sale and the destination has nothing to do with what you saved.
  • Unverified — the site refused the request. This is not the same as dead, and deleting these in bulk will throw away working bookmarks.
  • Alive but changed — no checker can catch this. The URL resolves, the server is happy, and the content is not what you saved. Only you can tell.

The part worth doing after

Finding the dead ones is satisfying and slightly beside the point, because the same thing will happen to the survivors. Every link in that list is one account deletion or one domain renewal away from joining them.

So the audit is worth pairing with a decision: which of these do you actually need to still have in five years? For those, a copy beats a link, because the ways a link dies are almost all things that cannot touch a copy you already hold.

Why Chrome will not do this for you

There is no built-in check, and that is unlikely to change. Verifying a bookmark means making a request to every saved URL, which is a privacy problem to automate quietly on a user's behalf and a performance problem to do at scale. So the job falls to you, or to a tool you point at an export.

Be aware of what a checker can and cannot tell you. A request that comes back 404 or fails to resolve is a confident dead. A 403, a 429 or a timeout often means the site is refusing an automated request rather than that the page is gone — Instagram, X and LinkedIn all do this routinely. A good checker reports those separately instead of declaring them dead, and anything in that category is worth opening by hand before you delete it.

Frequently asked questions

Does Chrome have a built-in broken bookmark checker?

No. Chrome offers no way to test whether saved bookmarks still resolve, and there is no setting or flag for it. You export your bookmarks and check them with a tool.

Is it safe to use an online bookmark checker?

Your bookmark file is a record of your interests and browsing, so it is worth caring where it goes. A checker that runs locally in your browser never uploads the file; one that asks you to upload it does. Prefer the former.

Why is a bookmark reported as broken when it opens fine for me?

Usually because the site refused an automated request — a 403, a rate limit or a timeout — rather than because the page is missing. Social platforms in particular block requests that are not a signed-in browser. Treat those as 'needs checking', not dead.

What should I do with the dead ones?

Delete the ones you cannot recover, try an archive for the few that matter, and for anything you want to keep going forward, store a copy of the content rather than a link to it.

Ready to clip your first post?

Save whole posts — text, images and video — from X, Instagram, TikTok, Reddit, YouTube and LinkedIn into one searchable library. Free and unlimited on your own device, with no card and no account.