Claude merges chat and Cowork memory to kill rebriefing
anthropic unified the memory systems for claude chat and cowork, so the ai now retains context across both environments to prevent repetitive rebriefing.
anthropic unified the memory systems of claude chat and claude cowork to eliminate the need for users to rebrief the agent when switching between research and execution modes. the update allows context to flow seamlessly between the two interfaces, enabling cowork to access details like project headcount or conference speakers discussed in chat. users maintain full control to read, edit, or delete memories, and the system updates memory dynamically during conversations. anthropic implemented default privacy guardrails excluding sensitive data like health or political views, though users can opt to include them.