1Password lets AI agents log in without seeing your password

1Password launched 1Password for Agents on Thursday. The product lets AI agents complete logged-in tasks without ever receiving the password they are logging in with.

The mechanics are specific. When an agent reaches a sign-in form, it sends a unique digital key to 1Password to verify its identity. The user receives an alert on their own device showing which credential the agent is asking for and why. If the user approves, 1Password fills the field directly rather than passing the password to the agent. The task completes; the credential stays in the vault.

The company's framing of the problem, as reported by TechCrunch: "Handing over a password, secret, or credential is a common shortcut to an agent, but can leave people with sensitive information in places they can no longer control as a copy gets saved for the next time."

The launch arrives wired into most of the agent landscape, including Anthropic's Claude, OpenAI's ChatGPT Work and dots, Meta's Muse, Grok Bot, and smaller agents from Town. It is included with eligible Individual and Family accounts on 1Password.com. The .eu and .ca accounts are not currently supported.

The timing is pointed. TechCrunch reports that U.S. banks have begun detecting agent activity inside customer accounts. Chase has told customers they "should not give their usernames and passwords to anyone."