1Password lets Claude sign in to websites without ever seeing the password

1Password launched 1Password for Claude on July 16, 2026, a browser integration that lets Anthropic’s Claude sign in to websites using stored credentials. The credentials are used at runtime without the secret ever reaching the model, its memory, or Anthropic’s systems, the company said.

The design targets a core problem in agentic browsing: how to let an AI agent act on your accounts without handing it your passwords. 1Password stays the source of truth for each secret, granting access only at runtime and injecting the credential directly into the target page for login and one-time-passcode tasks.

An “Agentic Mode” locks the 1Password vault down to only the credentials explicitly granted for the current task when an AI agent takes control of the browser. “The answer isn’t handing agents your secrets,” said Nancy Wang, 1Password’s chief technology officer. “It is to let a user give an agent permission to use a credential without letting the agent see it.”

The integration is available now for 1Password users on Mac across business, family and individual plans, and requires the 1Password desktop app and browser extension plus the Claude desktop app and the Claude in Chrome extension. Support for payment cards and personal details is planned after launch, and the feature works with Claude first, with 1Password saying it plans to expand to other AI agents.

The security model is 1Password’s own account and has not been independently audited; injecting credentials into live pages still depends on the browser and agent behaving as described. How it holds up against real agent workflows, and whether the promised expansion to other agents lands, will test the approach.