Check passwords & emails against 15B+ leaked credentials
🔒 k-anonymity for passwords✓ Not saved to a database✓ No registration
🛡Uses k-anonymity — your data is never sent in full.
Why Check for Breaches?
Billions of credentials are leaked every year. Attackers use these in credential stuffing attacks — trying leaked username/password pairs across thousands of websites. Even reusing a single exposed password puts all your accounts at risk.
📊
15B+
credentials leaked
⚡
1 in 3
passwords reused
🔓
81%
breaches use weak passwords
How the data breach checker works
Two checks that work differently
The checker has two tabs that answer different questions. The Password tab asks whether a password appears in Pwned Passwords, a collection of passwords exposed in past breaches. The Email tab asks whether an address appears in the breach records compiled by Have I Been Pwned (HIBP) or XposedOrNot. The two checks also handle your data very differently, so read the details below before deciding what to type.
Password check: a k-anonymity range query
Your browser computes the SHA-1 hash of the password locally with the Web Crypto API and writes it as 40 hexadecimal characters.
Only the first 5 characters of that hash are sent, directly from your browser to api.pwnedpasswords.com. InboxBurn's servers are not involved in this request.
The service replies with every hash suffix in its collection that starts with that prefix, each with the number of times it was seen. The request asks for padded responses, which adds dummy entries so the size of the reply reveals less about the prefix.
Your browser compares the remaining 35 characters of your hash with that list. A match gives the number of sightings; no match shows zero.
There are 16 to the power 5, roughly one million, possible prefixes, and each one is shared by many unrelated passwords, so the service only learns that your password belongs to a large group. Neither the password nor its full hash leaves your device.
Email check: what is sent and where
Email checks work differently: the full address is sent. Your browser passes it in a request to the InboxBurn API, which checks the format, verifies the Cloudflare Turnstile challenge when it is enabled, and forwards the address to the source you selected: the HIBP breached-account API, queried with InboxBurn's API key, or the XposedOrNot breach-analytics API. The reply lists each breach's name, domain, date, number of affected accounts, exposed data types and a short description.
The InboxBurn route does not save the address to its database. It is, however, transmitted to InboxBurn and then to the provider you chose, whose own privacy policy applies, and because it travels as a URL parameter it can appear in standard web-server request logs like any URL. The endpoint is limited to ten requests per minute per client.
HIBP and XposedOrNot maintain separate collections, so the same address can produce different results. Checking both gives a broader picture.
Interpreting the results
Password seen N times: this exact password is in breach data that attackers feed into credential-stuffing and dictionary attacks. Stop using it everywhere, however strong it looks. Even a count of 1 matters.
Password not found: it is not in this collection. That does not make it strong; a short or predictable password can be guessed quickly even if it has never leaked.
Email found: the address was part of a dataset exposed at the listed service. It does not necessarily mean your password was included, so look at the listed data types, such as email addresses, passwords or phone numbers.
Email not found: the selected source has no record. Many breaches are never disclosed or never reach these collections, and providers may omit certain breach categories from public lookups, so a clean result is not proof of safety.
Limitations to keep in mind
The input accepts printable ASCII characters and up to 128 characters. Accented letters, Cyrillic and other non-Latin characters, and emoji are removed as you type, so a password containing them cannot be checked accurately here.
Results reflect each provider's data at the moment you ask. A recent breach may not have been loaded yet, and old entries remain after you have changed a password. An email result does not say whether the leaked password is still valid, and an address can appear in breach data because someone else typed it when signing up.
If the HIBP source is not configured on the server, if a provider rate-limits the request, or if the captcha fails, the tool shows an error instead of a result. Try the other source or wait a moment.
Practical next steps
Change the password at the breached service and everywhere you used the same or a similar password.
Give every account its own random password; a password manager makes that practical.
Enable two-factor authentication, preferably with an authenticator app or security key rather than SMS where available.
Expect targeted phishing that mentions the breached service; leaked addresses are often reused for such campaigns.
Using separate addresses for low-trust sign-ups limits how far one leak spreads and shows which service exposed an address.