1Password for Agents now allows AI agents to log in to websites for you without seeing your password.
When an agent reaches a login page, it sends a request to 1Password through a unique digital key. The user receives a prompt on their device showing which credential the agent needs and why; from here, they can approve or reject the request.
Agents reportedly can’t approve access themselves. If a user approves the request, 1Password autofills the username, password, or one-time code directly into the sign-in page. The credential stays in the 1Password vault, and the agent does not see or store it.
1Password says it released this feature in response to seeing AI chatbot users paste their credentials into their chats.
“The old model where you hand over a credential and hope for the best doesn’t work,” chief product officer Aashish Tripathi said in a statement. “An agent should be able to do the job it’s been asked to do without getting permanent access to everything behind the account, and 1Password provides the trust layer that makes this possible across personal agent experiences.”
Users can manage agent access and disconnect an integration at any time. If they change a password in 1Password, connected agents can use the updated login—no other steps necessary.
1Password will support Meta Muse and SpaceXSI Grok Bot at launch. More integrations are coming soon, including Anthropic‘s Claude Duet and OpenAI’s ChatGPT Work. The company is also working on other integrations, like Browserbase and Browser Use, which both help AI agents access the web on users’ behalf.
1Password for Agents is included at no extra cost for Individual and Family accounts, though it’s not currently available for accounts on 1Password.eu or 1Password.ca.
