Cloudflare Adds Support for Claude Managed Agents

Cloudflare added support for Claude Managed Agents so developers can run and manage Claude agents within Cloudflare. Developers can connect agents to private systems, select a runtime environment, and monitor agent activity using Cloudflare services. Agent logic and orchestration remain on the Claude platform, while code execution and tool operations can run through Cloudflare’s infrastructure. The default deployment includes sandbox controls and logs, options for lightweight or VM-based execution, private service access, browser session monitoring, built-in email capabilities, and customizable tools. Support includes self-hosted sandboxes and MCP tunnels, allowing tools to run in developers’ own environments while keeping orchestration on Anthropic. Other providers also support the decoupled model.

"Until now, using Claude Managed Agents has meant running the entire stack on Anthropic-provided infrastructure. While this is great for some developers, others may need more control over their infrastructure choice, whether this is for security, compliance, or performance reasons."

"Cloudflare’s integration with Claude Managed Agents lets developers run agent logic on the Claude platform while using Cloudflare for code execution, secure connectivity, and custom tools. The default deployment includes sandbox controls and logs, support for lightweight or VM-based execution, private service access, browser session monitoring, built-in email capabilities, and customizable tools."

"Anthropic describes this model as “decoupling the brain from the hands.” The agent logic and orchestration remain in Anthropic, the “brain,” while the execution layer, the “hands,” can run on the cloud infrastructure of choice."

"They now support self-hosted sandboxes and MCP tunnels, enabling developers to run agent tools in their own environments while keeping orchestration on Anthropic."

Read at InfoQ