Cloudflare Brings Scalable Sandboxes to Claude Agents

Cloudflare Brings Secure, Scalable Sandboxes to Claude Managed Agents

New integration combines Anthropic’s intelligence with Cloudflare’s global network to provide a highly secure and scalable foundation for autonomous AI

Cloudflare announced it is collaborating with Anthropic to launch Cloudflare Environments for Claude Managed Agents. This integration allows organizations to run core agent loops on the Claude platform while using Cloudflare’s global network and Workers developer platform to execute code, secure private connections, and provide agents with specialized tools. By giving developers a secure and massively-scalable environment, Cloudflare is helping ensure that the next generation of AI assistants can be built quickly, securely, and at a global scale.

For AI agents to be truly useful in the workplace, they need to be more than just smart — they need to be secure and scalable. Cloudflare Environments for Managed Agents solves this by providing the foundation these agents need to work safely within an organization. A Workers-based control plane spins up secure sandboxes for every agent session, and security and compliance controls are applied by default. This enables businesses to maintain strict security protocols while scaling their AI initiatives from a single prototype to millions of users without infrastructure bottlenecks.

“The true potential of AI agents will only be realized when they can safely interact with the real world at massive scale,” said Matthew Prince, co-founder and CEO of Cloudflare. “By partnering with Anthropic, we are giving developers the ability to execute code and access private data securely—backed by the speed and scale of our global network. Now, businesses can focus on building the most innovative AI applications on the planet."

Cloudflare Environments for Claude Managed Agents enables developers to:

Scale globally with diverse runtimes: On Cloudflare Workers, developers can choose between full Linux-based microVMs for complex tasks or lightweight V8 Isolate-based sandboxes that offer millisecond boot times, allowing agents to scale quickly to meet massively-scaled workloads while keeping resource costs low. This makes agentic workloads possible for any large-scale consumer Internet company.
Secure sensitive data with Zero-Trust connectivity: Protect agents with advanced egress controls and Cloudflare Mesh. This allows agents to connect to internal, private services via post-quantum ready encryption, designed to keep sensitive data off the public Internet.
Audit agent behavior with native observability: With Cloudflare, Agents also come equipped by default with built-in tools to provide an audit trail and session recordings, ensuring actions the agent takes are secure and compliant.
Extend agent capabilities with customizable frameworks: Extend agent capabilities by writing simple functions directly on Cloudflare’s developer platform. This allows agents to create and execute their own tools on the fly within a secure, high-performance environment.

A Partnership for Secure AI Infrastructure

The launch of Cloudflare Environments marks the latest milestone in the ongoing collaboration between the two companies to solve the last mile of AI connectivity.

Source: Cloudflare media announcement

AI-driven orchestration for real-time BSS

Follow @PipelineWire