A professional checking an AI assistant's morning briefing on a phone

Anthropic has joined the race for the proactive AI agent. On Wednesday, October 21, the company unveiled Claude Duet, a standalone app that monitors your connected apps, sends you a morning briefing, and works on delegated tasks in the background — messaging you when something needs a decision, rather than waiting to be asked.

What Duet does

Duet is built around Claude Opus 5.5, Anthropic’s top model, and aimed squarely at professionals and small-business owners “with more on their plates than they can track.” Each morning it sends a short brief on the most important items across connected apps, then checks those accounts every few hours and alerts you only when something needs attention. Anthropic says it checks at intervals rather than constantly, because constant checking would burn through a subscriber’s usage limits much faster.

Hand it work in a single ongoing thread — by text or voice — and it can schedule meetings, draft customer follow-ups, build financial forecasts and slide decks, or create simple web apps. Tasks run on Duet’s own virtual machine with its own browser, so work continues after you close your laptop or switch off your phone. It carries over memory, skills, and connected apps from your existing Claude account, and reaches more than 6,000 connectors and plugins in the Claude Marketplace, including Gmail, Slack, QuickBooks, Stripe, Gusto, Docusign, HubSpot, Canva, and Shopify. Purchases can go through Stripe’s Link wallet and Shopify’s Shop Pay.

The approval-first pitch

The feature Anthropic stresses most is the permission model. Before Duet sends a message, makes a payment, signs a document, or deletes a file, it shows exactly what it is about to do and waits for your approval. Trading securities and moving money between accounts are off-limits even with approval, and a dedicated classifier is designed to reject requests to share credit card numbers or government IDs. Sensitive information such as health or personal financial details goes out only when the user has explicitly told Duet to send it.

Anthropic is candid about the limits: Duet can misjudge urgency, and it remains less effective at personal tasks such as booking tickets. That honesty matters, because the whole product rests on a claim no lab has yet demonstrated at scale — that an agent holding standing access to your inbox, calendar, and payments asks at the right moments, and not at all the others.

A crowded starting line

Duet arrives weeks after its two obvious rivals: OpenAI’s always-on Dots agents, which pursue user goals across apps like Slack and Microsoft Teams, and Meta’s Muse agent, which drew millions of downloads. All three share the same bet — that the next interface is not a chatbot you open, but an agent that comes to you. Anthropic’s differentiation is narrower and more deliberate: a business-chores agent with visible guardrails, aimed at the people most likely to pay for it.

Availability

For now, Duet is a research preview on a waitlist, limited to US subscribers on Claude Pro and Max plans, sharing their existing usage limits. Team and Enterprise plans are not included, and Anthropic has announced no Duet API or enterprise rollout schedule. It is available as a standalone iOS and Android app and on claude.ai.

Whether professionals trust an agent that reads their inbox before breakfast will decide whether Duet is a product or a preview. The waitlist is the first test.

Sources: Reuters (Oct 21, 2026); MacRumors (Oct 21, 2026); VentureBeat (Oct 21, 2026). Product details are Anthropic’s own claims and await independent testing.