Claude Duet takes aim at Dots, Grok Bot and Muse
The post breaks down Claude Duet: it inherits Claude memory and connected apps, runs on Opus 5.5, keeps working while you're away, and ships with safety limits, framing agents as a full product surface for Anthropic.
we go to sleep and anthropic ships another agent. claude duet is a proactive one that surfaces what actually needs you, keeps working while you’re away, and only asks when it matters. direct shot at dots, grok bot, hark, muse. it inherits your claude memory + connected apps and runs on opus 5.5 for the judgment calls. safety is baked in (no password access, classifier on sensitive stuff, final say stays with you). morning brief, app monitoring, task completion, reports, even building things. early research preview but the product shape is clear. anthropic is treating agents as a full product surface now, not just a model feature.
This post quotes another post: x.com/claudeai/status/2112984664256241837