Anthropic wants to own your agent's memory, evals, and orchestration — and that should make enterprises nervous

Anthropic expanded Claude Managed Agents into a more complete enterprise agent platform.

New features include Dreaming for agent memory, Outcomes for built-in evaluation, and Multi-Agent Orchestration for task delegation.

The update folds tools enterprises often use separately into Anthropic’s hosted stack, simplifying deployment.

It may also raise concerns about vendor lock-in, compliance, and dependence on one provider for critical workflows.