Anthropic Launches Self-Hosted Claude Managed Agents in Private Beta

• •Anthropic's self-hosted Managed Agents are now in private beta, enhancing control for enterprises.
• •The setup allows companies to run the entire agent framework on their own infrastructure.
• •Security measures ensure that sensitive credentials are managed outside the agent's sandbox.

On November 12, 2026, Anthropic announced the private beta release of self-hosted Claude Managed Agents, allowing companies to run the entire agent framework within their own cloud or on-premise environments. This includes the agent harness, sandboxes, credential management, and session history, which previously required Anthropic's infrastructure. The self-hosted setup enables organizations to maintain greater control over their internal systems while using autonomous agents for tasks such as fixing broken builds and triaging alerts. Customers can now operate the Managed Agents server, PostgreSQL database, and the workers that execute agent tasks. The service is designed with security principles to ensure that credentials remain protected outside the sandbox environment. Organizations can configure their environments to restrict agent access to only approved hosts, enhancing security. This release responds to the growing demand for more autonomous agents capable of handling work within corporate systems.