Agent Evaluation

1 article

Editorial pen-and-watercolour schematic of a Claude Managed Agents system: a lead agent carrying the Claude wordmark delegating to specialist sub-agents on a shared filesystem, with a separate grader checking output against a rubric and a dreaming memory-curation loop
Claude Launch AnalysisClaudeAnthropic

How Claude Managed Agents Actually Work: Dreaming, Outcomes, Multiagent Orchestration, and Webhooks (2026)

Anthropic gave Claude Managed Agents four new mechanics at Code w/ Claude: Dreaming, Outcomes, Multiagent Orchestration, and Webhooks. The one that changes how you build is Outcomes — a separate grader that loops the agent until a rubric is met. Here is how each one works, and when to reach for it.

Marco Lobo
Marco Lobo·25 May 2026·9 min read