Anthropic’s Claude Managed Agents can now “dream,” sort of
Summary
Anthropic introduced a feature called 'dreaming' for Claude Managed Agents, a scheduled process that reviews past sessions and memory stores to identify high-signal information to preserve for future tasks. Dreaming supports cross-agent memory and long-running multi-agent orchestration and is currently in research preview restricted to Managed Agents on the Claude Platform. Anthropic also noted broader access to related features and is doubling five-hour usage limits for Pro and Max Claude Code subscribers.