Cloudflare Sandboxes Execute Claude Managed Agents
Run the agent loop on Anthropic's platform while offloading code execution, private service access, and tool calls to Cloudflare Sandboxes—decoupling the brain from the hands.
May 20, 2026
Summary
Developers can now choose execution infrastructure independent of the agent orchestration layer, enabling compliance-driven architecture, cost optimization via lightweight isolates instead of full microVMs, and zero-trust credential injection without exposing secrets to sandbox environments.
Why it matters
Developers can now choose execution infrastructure independent of the agent orchestration layer, enabling compliance-driven architecture, cost optimization via lightweight isolates instead of full microVMs, and zero-trust credential injection without exposing secrets to sandbox environments.
Implementation verdict
This replaces running the entire Claude Managed Agents stack on Anthropic-provided infrastructure. Requires forking the provided deployment template and configuring backend type (microVM or isolate). Ready now—Cloudflare provides an onboarding guide and default template for minutes-to-deployment setup.
Sources
- 1.decoupling the brain from the hands
- 2.The core agent loop runs in Anthropic (the "brain"), but the infrastructure for running and executing code (the "hands") can be run anywhere, including Cloudflare
- 3.If we're constantly running a full microVM per agent, we'll be unnecessarily burning a ton of resources and money to enable this scale
- 4.Claude can read files, run commands, browse the web, and execute code
- 5.This lets you inject secrets into requests outside the sandbox, so the agent never has access to them
Dev Signal
Get briefs like this in your inbox — free, 3x a week.
100+ sources compressed into one 4-minute read. Ranked, cited, implementation-ready.