Claude Managed Agents at a glance
Claude Managed Agents is a set of APIs for running cloud-hosted Claude agents, with Anthropic handling the harness, sandboxes, permissions and long-running sessions.
What changed with Claude Managed Agents
Anthropic launched Claude Managed Agents in public beta on the Claude Platform. It packages a managed agent loop with sandboxed code execution, authentication, tool execution, session tracing and scoped permissions, so agents can run for hours and keep progress across disconnections.
Pricing is consumption-based: normal token rates plus $0.08 per session-hour of active runtime. Multi-agent coordination and an Outcomes feature, where Claude checks its work against success criteria, started as research previews. Memory, scheduled deployments and self-hosted sandboxes were added in later updates.
Why Claude Managed Agents matters for businesses
For agencies, this shifts some of the plumbing of agent projects to the model provider. The trade-off is closer coupling to one vendor's runtime, so it is worth weighing against self-hosted frameworks for each client.
Put it to work: AI Agents
Sources for this story
- Claude Managed Agents: get to production 10x faster (opens in a new tab)Official
Anthropic · claude.com
- Claude Platform release notes (opens in a new tab)Official
Anthropic · platform.claude.com
Last verified: . IA Compa summarizes public announcements for business readers. We are not affiliated with the companies mentioned. Always confirm details, pricing, and availability with the official source.
- #agents
- #anthropic
- #claude
- #api
- #managed-agents
