Anthropic opens private beta for self-hosted Claude Managed Agents
Anthropic's self-hosted Claude Managed Agents private beta lets customers run the agent harness, sandboxes and session history in their own cloud or on-prem.

Watch the short
Under a minute, read by an AI anchor. Made .
YouTube loads only when you press play.
Anthropic has opened a private beta of a self-hosted version of Claude Managed Agents. It lets customers run the agent harness, the sandboxes, and the session history in their own cloud or on-premises, and choose where Claude requests are sent. The announcement came from the ClaudeDevs account.
Katelyn Lesse's post adds four features to the release: agent identity, OpenTelemetry support, kill switches, and an evals product.
What changed
Anthropic opened a private beta of a self-hosted Claude Managed Agents that lets customers run the agent harness, sandboxes and session history in their own cloud or on-premises, while model requests still go to an external provider.
Type: New capability · Applies to: Claude Managed Agents
Why it matters
Companies with data-residency or security constraints could keep agent execution and session records inside their own infrastructure while still calling Claude.
What you can do
If data residency matters, request access to the private beta and map which agent data would stay on your side.
How the story developed
Adds that model queries still route to an external provider, Claude does not run inside the customer's network, and webhooks and scheduled runs are not yet available.
Original source
We have no link to the original source for this story yet.
Further reporting
Where we saw it
Latent Space, The AI Report
AI wrote this summary from the sources listed, and nobody reviewed it individually before publication. Check the original source before you rely on it. How it works