prahapp
English
Free consultation
Menu

Products ·

Anthropic launches Claude Duet, an AI agent that reaches out before you ask

On 21 October Anthropic opened a research preview of Claude Duet. Built on Claude Opus 5.5, the agent watches your connected apps, sends a morning brief and works on tasks in its own cloud computer.

Duet is a standalone app for the web, iOS and Android. Unlike a normal chat it does not wait to be asked: every morning it sends a brief on what matters across connected tools, checks them again several times a day and only notifies you when something needs attention. Everything happens in one continuous thread, so the agent remembers what it has been working on.

Tasks run in a cloud virtual machine with its own browser, so your own computer can stay off. Anthropic says it handles anything from scheduling to dashboards, financial forecasts and web app prototypes. It can use more than 6,000 connectors and plugins, including Gmail, Slack, QuickBooks, Stripe and Shopify.

Safety is built around approval. Before sending a message, making a payment, signing a document or deleting anything, Duet always asks. Some actions, such as trading securities or moving money between accounts, are blocked even with approval. According to Anthropic the agent never sees passwords or sign-in tokens, and a dedicated classifier blocks sharing card and ID numbers.

Access is by waitlist, for now only for Pro and Max subscribers in the United States. There is no separate price: Duet shares usage limits with the rest of Claude, and background checks count toward them. The launch answers similar personal agents from Meta (Muse) and OpenAI (Dots).

What it means for companies

A proactive agent that watches email, invoices and orders on its own, and leaves anything with consequences to a person, is exactly the model we build for companies in Czechia. The difference is that a custom AI agent runs on your systems, with data in the EU, and does only what you allow. We explain how to keep such an agent in check in our article on AI guardrails. We build a working prototype for one of your processes free of charge within 3 working days.

Sources: Anthropic: Introducing Claude Duet, MacRumors

Bring us one process or one app idea.

In 30 minutes we tell you whether it is worth handing to an AI agent. Then you get a free working prototype within 3 working days. If you don’t like it, you don’t pay.

  • We reply within 24 hours on working days
  • The engineer who designs it also builds it
  • Write in Czech, English, German, Italian or Turkish

Contact

Write to us

Describe the process or app you want done. We reply within 24 hours on working days, and the prototype within 3 working days is free.