1Password lets AI agents log in for you without seeing your password

1Password launched 1Password for Agents, a way to let an AI agent sign in to your accounts without ever handling your credentials. When the agent needs to log in, it sends a unique digital key to 1Password to prove its identity. The user gets an alert and can approve or deny the attempt. Once approved, 1Password signs the agent in and fills the credentials itself, so the password never passes through the agent. The company's argument is that handing an agent your password leaves copies of sensitive information in places you can no longer control, and this keeps the user in the loop on every login.

The launch names a set of agents it works with, including Anthropic's Claude Duet, OpenAI's Dots and ChatGPT Work, Grok Bot, Meta's Muse, and agents from the startup Town, with more companies to come. The timing is pointed. TechCrunch notes an Axios report that US banks have begun detecting agentic activity in customer accounts and mostly cannot tell whether it reflects a genuine user request, with bankers calling the situation "horrifying" and a Chase spokesperson repeating that customers should not give their username and password to anyone.

The announcement is thin on the practical details that decide adoption. There is no pricing, no rollout timeline, and no plan tiers beyond the launch.

Why it matters

If you are letting agents act on your accounts, the password-sharing shortcut is exactly what banks are now flagging as a problem, and this gives you a way to delegate access while keeping an approval step and never exposing the credential. Whether it is worth adopting depends on details 1Password has not published yet, so the thing to watch is which services and agents actually support it.