Anthropic launches Claude Managed Agents in public beta

Anthropic launched Claude Managed Agents in public beta — a fully managed agent harness that runs Claude as an autonomous agent on Anthropic’s infrastructure.

What it is

Managed Agents provides a pre-built, configurable agent harness with:

• Secure sandboxing — isolated execution environments
• Long-running sessions — agents can work for extended periods
• Multi-agent coordination — parallelize complex work (research preview)
• Governance and execution tracing — identity management and audit trails
• Server-sent event streaming — real-time progress updates

Pricing

$0.08 per session hour plus standard API token costs. No fixed subscription or infrastructure cost.

Early adopters

Notion and Asana are among early adopters, with Anthropic claiming 10x improvements in time to ship.

API access

Currently in beta — all endpoints require the managed-agents-2026-04-01 beta header. Multi-agent coordination and memory features are in research preview and require access requests.