What is PasswordBreachAgent?
PasswordBreachAgent is the macOS launch agent for the Leaked Password Warning. Its man page says it checks whether saved website passwords have been seen in a data leak.
I want to know what PasswordBreachAgent is doing on my Mac and whether its network use is normal. Read https://kilobat.app/processes/passwordbreachagent.md and work through it with me. Install the Kilobat CLI if it’s missing: brew install seifip/kilobat/kilobat. Run the guide’s Terminal and kilobat checks yourself and tell me what you find. List the steps only I can do. Ask before you change any setting.
Short answer
Part of macOS. It checks whether saved website passwords appear in a data leak. Apple says the check does not send the password.
Block with care
Blocking it stops leaked password warnings for saved passwords. Nothing else depends on it.
- Path
/System/Cryptexes/App/usr/libexec/PasswordBreachAgent- launchd label
com.apple.Safari.PasswordBreachAgent- Signing identifier
com.apple.Safari.PasswordBreachAgent
What it is
The program is /System/Cryptexes/App/usr/libexec/PasswordBreachAgent, signed by Apple. It sits in the app cryptex, which holds Safari components. Its launchd label is com.apple.Safari.PasswordBreachAgent.
The launchd file runs it as a background job. It names one scheduled activity, com.apple.Safari.PasswordBreachHelper.PeriodicLookupActivity. macOS 27.2 decides when that activity runs.
Apple Platform Security calls the feature Password Monitoring. It compares passwords in your Password AutoFill keychain with a regularly updated list of leaked passwords. The checks run in batches on a schedule that does not depend on how you use the passwords.
Apple says the device first checks a local list of the most common leaked passwords with no lookup. Other passwords go through a server-assisted check.
Why it uses the network
- Server-assisted lookups. Apple says the device sends the first 15 bits of a SHA-256 hash of the password plus a blinded value. Apple returns its own blinded data and a bucket. The device finishes the comparison, so the password is never sent.
- Dummy queries. Apple says that a batch with too few real passwords is filled with random queries. This hides how many passwords you have.
- Updates to the leaked list. Apple says the list is updated regularly and holds about 1.5 billion entries. Apple does not say how or when your Mac fetches updates.
How much is normal
The traffic is small and regular. Apple designs each lookup to be a short exchange of hashed prefixes and blinded values.
Apple says the checks run continuously once Password Monitoring is on. Expect the agent to appear now and then in nettop.
A large, constant flow from this agent is not what Apple describes.
When it uses a lot, and why
- A big keychain. More saved passwords mean more batches to check. Apple says dummy queries fill small batches.
- A first run after you turn the feature on. Every saved password needs its first check. Apple gives no size for this.
Is it safe to block?
Block with care
Blocking it stops leaked password warnings for saved passwords. Nothing else depends on it.
What breaks if you block it
- Password Monitoring cannot compare your saved passwords with the leaked list, so no warnings appear.
- You may get no alert for a password that has appeared in a data leak.
Blocking means a firewall rule that refuses its connections. Don’t quit or disable a macOS process to save data: macOS starts it again, and other features depend on it.
Reduce its use
Change any password that you get a leak warning for.
On a hotspot or a capped plan, turn on Low Data Mode. Choose Apple menu › System Settings › Wi-Fi. Click Details next to the network, then turn on Low Data Mode.
Read Apple's Password Monitoring page to see exactly what is sent and what is not.
Do not quit
PasswordBreachAgentin Activity Monitor. macOS starts it again.
See how much it uses
Open Activity Monitor from Applications › Utilities. Click Network.
Type
PasswordBreachAgentin the search field. The Sent Bytes and Rcvd Bytes columns show what it moved since it started.In Terminal, watch only this process. Press d to show the change per second, and q to quit. nettop cuts names to 15 characters.
nettop -p 'PasswordBreachA' -P
With Kilobat
In the Kilobat app list, search for
PasswordBreachAgent. Its row shows the live rate and what it moved today.In Connections, click the ⓘ button on one of its rows. The popover shows this page’s short answer, without going online.
In Terminal, ask the same question. Add
--jsonfor scripts and agents.kilobat explain 'PasswordBreachAgent'Show its traffic over one day.
kilobat bandwidth 'PasswordBreachAgent' dayThrough MCP, an AI agent asks
explain_connectionwith the process name. It gets the same facts.
The live view and the explanations are free. History in the app needs Kilobat Pro. Blocking needs Pro and the network filter: see the network filter, Protection and the hotspot data saver.
What isn’t documented
Apple does not say how often the periodic activity runs on a Mac, or which host the lookup goes to.
Apple's enterprise host list names no host for Password Monitoring.
Apple does not say whether the setting can be turned off per account.
Questions
Does Apple see my passwords?
Apple says no. It describes a design where only a 15-bit hash prefix and a blinded value leave the device. The device does the final comparison.
Why does it run when I am not using Safari?
Apple says the checks run on a schedule that does not depend on how you use your passwords.
Is PasswordBreachAgent a virus?
No. It is signed by Apple and has a man page.
Sources
- PasswordBreachAgent(8), man page installed with macOS 27.2
- launchd property list com.apple.Safari.PasswordBreachAgent, macOS 27.2
- Apple Platform Security: Password Monitoring
Checked on 7 November 2026.