DigestAI news desk

Cut through the AI noise.

Marketing & Small Businessupdated 5 min read

Anthropic opens Claude Duet research preview for US Pro and Max users

On October 21, 2026 Anthropic opened a research preview of Claude Duet, an app that turns its Claude model into a proactive agent. Access is limited to US users on the Pro and Max plans via a waitlist. The app is available as a standalone iOS and Android app and on claude.ai, using the same usage limits as the rest of Claude.

3 sources

What you can do with it

AI at Work →
Claude Duet by Anthropic

Schedule meetings and draft invoices, directly in Claude Duet

Monitors email and calendar, then schedules meetings or drafts invoices after you approve.

Who for
marketers
Cost
Price not stated
Effort
minutes
Already included in
Claude Pro and Max plans

Use it for

  • Schedule meetings
  • Draft invoices
  • Monitor calendar

Watch out Background activity counts toward usage limits

Claude Duet in the tool directory

Key points from the news

  • Claude Duet research preview launched for US Pro and Max users, accessed through a waitlist.
  • Built on Claude Opus 5.5, the app runs on its own VM, checks apps every few hours, and uses shared usage limits.
  • Supports over 6,000 Claude Marketplace connectors; requires user approval for payments, signatures, and blocks trading securities.

Claude Duet is built around Claude Opus 5.5 and runs on a dedicated virtual machine with its own browser, so it can act without the user’s device being on. It checks connected apps every few hours, counts toward plan limits, and sends brief morning summaries and alerts when tasks need attention. Users interact via typed or voice messages in a single thread.

The agent can schedule meetings, build dashboards, draft financial forecasts, prototype web apps, and integrate with over 6,000 Claude Marketplace connectors such as Gmail, Slack, QuickBooks, Stripe, Gusto, LinkedIn, Canva, and Shopify. Before any payment, document signing, or deletion, Claude Duet shows the proposed action for user approval; trading securities and moving money between accounts are blocked. Anthropic says safety is handled by a dedicated classifier, red‑team testing, and a HackerOne bug bounty. Early testers include Anthropic staff and a handful of small‑business owners.

Model page: Claude Opus 5.5 →

Full story from Unite.AI · by Jonas Reeve, Cognitive AI & AGI, AI Research AgentOpen source ↗

Anthropic Opens Claude Duet Research Preview to US Pro and Max Users

Unite.AI · 21 October 2026

Loading the full article…

This text was published by Unite.AI and written by Jonas Reeve, Cognitive AI & AGI, AI Research Agent. It is reproduced here with attribution so you can read it in full; the rights remain with the publisher. Read it at the source ↗

Coverage and discussion

3sources
Topics · follow one to build your own front page
AnthropicClaude Opus 5.5Boris ChernyChelsea LarssonKaran AhujaJess ProvencherAmin EbrahimiMya Dalton

The headline, key points and digest above were generated by Digest AI's editorial model from the linked sources. Automated summaries can contain errors: the sources are the record. Spotted a mistake? Tell us. Published by Martin K., who runs Digest AI and handles corrections.

Comments

via GitHub Discussions

More in Marketing & Small Business

All →

Related stories