返回资讯中心
外部精选
软件工程
#开发者实时流

1Password adds browser login feature for AI agents

The 1Password for Agents platform integrates with AI agents like Muse, from Meta, Anthropic's Claude, and OpenAI's ChatGPT Work and dots, to keep your credentials and secrets secure, while your agents get thin…

Hacker News 最新walski2026/10/22 · 23 天前8 分钟阅读

以下正文同步自 Hacker News 最新,版权归原站所有,已转换为易读排版。

The 1Password for Agents platform integrates with AI agents like Muse, from Meta, Anthropic's Claude, and OpenAI's ChatGPT Work and dots, to keep your credentials and secrets secure, while your agents get things done.

Give your agent a safe way to use your Instacart or DoorDash to order your groceries.

Securely give your agent access to Canvas and Google to add exams and assignments to your calendar.

Let your agent log in to your airline and Airbnb accounts to book your next trip out of town.

Vibe code an app or website in GitHub and save API keys in your vault.

1Password will connect to your personal agents, whether you have one or multiple, so they can work without seeing your passwords. Join the waitlist to be notified when your agent is ready.

Whether you have one AI agent or multiple, you’ll be able to connect 1Password to any or all supported agents. 1Password for Agents is included in all Individual and Family plans.

Rolling out over the coming days across all platforms.

Rolling out over the coming days across all platforms.

Even when you approve access to an account, your AI agents don’t see your passwords. Everything in your vault remains protected.

No. 1Password for Agents lets AI agents sign in without receiving or seeing the credential. When an agent reaches a sign-in form, it uses a unique digital key to verify its identity to 1Password. The person sees which credential it needs and why, and then chooses whether to approve the request. That approval happens on the person’s own device, confirming that the permission came from them. 1Password then fills it directly into the sign-in form instead of sending it to the agent. Once signed in, the agent continues the task while the password stays protected in 1Password.

No. Connecting an agent does not give it access to everything in your vault. It requests a login for a task, and you decide whether to approve that request. The agent can’t see the credentials, and your other saved passwords in your vault remain protected by end-to-end encryption.

Yes. You review the access request and can approve or deny it. If more than one saved login matches, you can choose which one 1Password uses. How often you are asked to approve access may vary by integration.

Agentic Autofill helps a supported agent sign in through a browser without handling the password itself. When the agent requests a login and you approve it, 1Password fills the password or one-time code into the sign-in page so the agent can continue its task.

An agent may be able to see information or take actions within the signed-in account. Its behavior there depends on the agent and the website. Review the task and the account before approving access.

1Password for Agents works with supported personal and coding agent integrations, rather than every AI tool. Check the integration cards on this page for available and coming-soon connections. Features may vary by agent.

The new personal agent integrations are being introduced for 1Password Individual and Families accounts. Availability and setup requirements can vary by integration, region, device, and sign-in method. Check each integration’s details before getting started.

No. These integrations are optional. If you do not connect an agent, you can continue using 1Password as you do today.

正文由 探How 从来源站点 RSS 同步,内容未经改写;遇到排版缺失或需要图片、视频时请以原文为准。

打开原文
去社区讨论 →