Anthropic previews Claude Duet, a proactive agent app
Original title: Claude Duet
The iOS, Android, and claude.ai preview is in Pro and Max plans, built on Claude Opus 5.5, and waits for approval before sending, paying, signing, or deleting.
Anthropic opened a research preview of Claude Duet, an app that messages users about connected accounts, keeps working in a virtual machine when they step away, and asks before acting. It is on the waitlist for Pro and Max subscribers and uses the same Claude usage limits. Credentials stay out of the agent’s direct view, and a new classifier is meant to block leaks of credit card numbers and government IDs unless the user explicitly allows other sensitive data.
- It checks connected apps every few hours and sends a short morning brief.
- Trading securities and moving money between accounts stay off-limits even with approval.
- Users can disconnect any connected app.
- Red teams and an external firm tested it, and it is in Anthropic’s public bug bounty.
- The preview carries over memory, skills, and connected apps from the user’s Claude account.
- Vendors
- Anthropic
- Products
- Claude Duet
- AI models
- Claude Opus 5.5
From the original article
6 min read at claude.comToday, we’re opening a research preview of Claude Duet, a new app where Claude works as a proactive agent. It comes to you with what needs your attention, completes tasks you ask of it, keeps working even when you step away, and checks with you before it acts on your behalf. We built Claude Duet for professionals and small business owners with more on their plates than they can track.
By zaevlad · claude.com
A short excerpt; the article and its images belong to the publisher. Summary, key facts and assessment above are AI-written from the article and can be wrong.