ChainCatcher report: Anthropic has announced the open private beta of Claude Managed Agents for self-hosting. Enterprises can now deploy the AI Agent’s execution framework, sandbox environment, access credentials, and session history on their own cloud servers or on-premises data centers, eliminating the need to entrust these components to Anthropic. In May of this year, Anthropic already allowed enterprises to self-host the sandbox, enabling AI to run code on internal servers—but the core systems responsible for agent orchestration, model invocation, and session logging remained under Anthropic’s management. The new update brings these control systems entirely inside the enterprise’s internal network. When using Agents to automatically fix code, test software, or troubleshoot system issues, companies can restrict the servers they can access, manage API keys, and log and audit every step via OpenTelemetry. The Claude model itself still runs in the cloud; during agent reasoning, prompts, code, and tool outputs must still be sent to the model provider. Enterprises may choose to invoke Claude through Anthropic, Amazon, Google, or Microsoft cloud services, or route all model requests through their own servers first to centrally control access permissions, invocation channels, and costs. This feature is currently available only in a private beta for approved enterprises and requires contacting Anthropic’s customer team to obtain deployment files.
Anthropic Launches Self-Hosted Private Testing for Claude Managed Agents
ChaincatcherShare
Summary
According to on-chain news, Anthropic has launched a private test for self-hosted Claude Managed Agents. Enterprises can now install the agent framework, sandbox, and credentials on their internal servers rather than relying on Anthropic’s hosting. Previously, core components such as agent scheduling and storage remained on Anthropic’s cloud. The new version enables companies to control access and use OpenTelemetry for logging. While Claude still runs in the cloud, organizations can route model requests through their own servers or major cloud providers. The feature is currently available only to approved businesses and requires direct contact with Anthropic for deployment. Inflation data and infrastructure updates are driving increased interest in AI infrastructure development.
Source:Show original
Disclaimer: The information on this page may have been obtained from third parties and does not necessarily reflect the views or opinions of KuCoin. This content is provided for general informational purposes only, without any representation or warranty of any kind, nor shall it be construed as financial or investment advice. KuCoin shall not be liable for any errors or omissions, or for any outcomes resulting from the use of this information.
Investments in digital assets can be risky. Please carefully evaluate the risks of a product and your risk tolerance based on your own financial circumstances. For more information, please refer to our Terms of Use and Risk Disclosure.