Password Breach Checker
See whether a password has appeared in a known data breach, without the password leaving your device. 100% free, no signup. Everything runs in your browser.
Loading the tool…
A password that has appeared in a breach is a password that is on a list, and the lists are what attackers try first. Not clever guessing, not brute force from a to z, just a few hundred million strings that have already worked somewhere, run against every login they can reach. The uncomfortable part is that a password can be long, odd and personal to you and still be on that list, because someone else thought of it first, or because the site you used it on in 2014 was breached and nobody told you. I built this so you can find out in two seconds without doing the one thing you should never do, which is type a password into a stranger's website.
The trick that makes that possible is called k-anonymity, and it is worth understanding because it is the difference between this tool and a phishing page. The password is hashed here, in your browser. Of the forty characters in that hash, five are sent to Have I Been Pwned's range service. Those five identify a bucket of several hundred unrelated hashes, which come back in full, and the match is found on your device. The service never sees the password, never sees the full hash, and cannot tell which of the hundreds you were interested in. The page shows you the exact five characters before you press the button. If the password turns out to be leaked, the password generator makes one that is not.
How to use
- Type the password you want to check. The field is hidden by default; the Show button reveals it if you want to confirm you typed what you meant.
- Look at the hint under the field. It shows the five characters that will be sent, so there is no mystery about what leaves your device.
- Press Check against known breaches.
- Read the count. Zero means it has not appeared in any breach the service knows about. Any other number is how many times it has.
- If it has appeared, change it everywhere it is used, starting with email and banking, and never reuse it.
Why use our password breach checker?
The password does not leave your device, and you can watch that being true. The hash is computed in the browser with SubtleCrypto. Five characters of it are sent; the other thirty five never are. The page prints those five characters before you click, and the network tab of your browser will show one request to a range endpoint and nothing else. The whole protocol is published by Have I Been Pwned, and it is why a security-conscious person can use this tool and should refuse to use one that asks for the password in plain text.
It answers a different question from a strength meter. Strength meters guess at entropy. This checks against reality: has this exact string been in a leak. A password can score well on a meter and have been leaked millions of times, because meters reward length and symbols and cannot know that a particular long string with a symbol in it was used by half a million people. Both questions are worth asking, and this is the one most sites skip.
The count tells you how urgent it is. Found once might be one obscure breach. Found fifty million times means it is in the first thousand lines of every cracking dictionary in existence. The page translates the number into plain language, and the honest answer is that any count above zero means change it, with the larger counts meaning change it today.
Padding hides even the size of the answer. The request asks the service to pad its response with fake entries, so the number of bytes coming back does not reveal how common the prefix is. A tool built for show would skip that detail; the protocol includes it for a reason.
Checking is not the same as fixing. A clean result on a weak, reused password is not a clean bill of health. The combination that actually protects an account is a password that is long, unique to that account, and stored in a manager so you never have to remember it. Once you have one from the password generator, there is nothing to check, because a random 16 character string has never been in a breach and will not be.
Who is this tool for?
The password you have had since university. Most people carry one or two old favourites across years of accounts. Those are precisely the ones that have had the most chances to leak. Check them, and if they come back with a count, treat it as a list of every account they protect that needs a new one. The text encryptor is one way to keep that list somewhere safe while you work through it.
Choosing a new password by hand. If you insist on inventing one rather than generating it, check it before you use it. A surprising share of hand-made passwords that feel unique are not, because people converge on the same patterns: a word, a year, an exclamation mark.
Helping someone who is not technical. A parent, a client, a colleague who has been told their password is weak and does not believe it. A count of several million is more persuasive than any lecture, and watching the hash prefix on screen is the fastest explanation of why this check is safe when the email asking them to confirm their password was not.
Auditing a small team. Not by collecting passwords, which you should never do, but by sending this page and asking everyone to check their own. Nobody has to reveal anything, and the ones who find a hit know exactly what to change.
Frequently asked questions
The password never leaves your device: hashed in the browser, and only the first five characters of the hash are sent, which identifies a group of several hundred unrelated hashes. The match is found locally. The page shows the exact five characters before you press Check, and your browser's network tab will show that one request and nothing else. The text encryptor works on the same principle: the secret stays with you.
Have I Been Pwned's Pwned Passwords service, which holds billions of passwords collected from public data breaches, the same dataset many password managers and sign-up forms check against, queried through the range API that is designed so the service never learns the password.
Change it on every account that uses it, starting with email, because email resets everything else. Use a different password for each account from now on, generated rather than invented, and store them in a password manager. If two-factor authentication is available on the account, turn it on.
No. It means it has not appeared in a breach the service knows about. A short or guessable password can pass this check and still fall to a dictionary attack. Strength and leak status are separate questions, and the password generator answers both by producing something long, random and never used before.
Because it tells you which list the password is on. A count of one or two means a single breach. A count in the millions means it is among the most common passwords on earth and is tried against every login within seconds of an attack starting. Any count above zero means change it; the big counts mean change it first.
No, and it is the one tool on this site that does not. The check requires one request to the breach database. What makes that acceptable is what the request contains: five characters of a hash, which reveal nothing about the password itself.
Not here. This tool checks passwords only. Have I Been Pwned's main site lets you search by email address to see which breaches an account appeared in, which is a different and equally useful question.

