1Password Now Lets Claude Log In For You. It Never Actually Sees Your Passwords.

Artificial intelligence is starting to do more than just answer questions. These days, AI agents can handle chores for you like shopping online or updating your accounts. The catch has always been security, because handing your login details over to a machine is a risky move.

Now, Toronto-based 1Password has built a way around that problem, and it’s designed specifically for Anthropic’s Claude AI chatbot.

The new integration lets Claude get through tasks that need a website login or a one time passcode, without the AI ever storing or seeing your credentials. When Claude runs into a login screen, 1Password steps in and asks for your permission, usually with a fingerprint or biometric check. Once you approve, the password manager fills in the details right on the page. Claude gets past the login to finish the job, but your info stays locked in your vault.

“We need a new security model that is purpose-built for agents, not just humans,” said Nancy Wang, CTO of 1Password. “The answer isn’t handing agents your secrets. It is to let a user give an agent permission to use a credential without letting the agent see it. Claude knows it used your login; it does not need the password or one-time code in its context. That distinction is where trust in agents starts and the foundation we’re building with Anthropic.”

To keep the AI from poking around your other saved passwords, 1Password also added a feature called Agentic Mode. The moment an AI takes control of your browser, the 1Password extension locks itself down. From there, Claude can only use the specific login you just approved for that one task.

For now, the tool is available for Mac users and needs both the 1Password and Claude desktop apps, plus their browser extensions.

Check out the video below to see it in action: