By Vero SecurityBuilt and funded by Vero Security, the AI Security Engineering platform.Read the charter

Anthropic sandbox-runtime

Anthropic open sourced the sandboxing layer behind Claude Code as a tool that wraps any process with filesystem and network restrictions. You can use it to contain other agents, MCP servers or scripts under the same policy model. It is also a practical way to learn how proxy-based network filtering works.

Read in this order

  1. Open sourceanthropics/sandbox-runtimeStart here: the tool itself, with config examples you can run today.
  2. ArticleMaking Claude Code safer with sandboxing (Anthropic)The design thinking behind it, including the network proxy approach.

Save your progress

We email you a sign-in link. No password.

By signing in you agree to the terms. The privacy policy explains how we use your details.

Join the AI security engineering community

Free. Roadmap updates, chapters near you, and first word when credentials open.

We’ll email you community updates. Unsubscribe anytime. By sending this you agree to the terms. The privacy policy explains how we use your details.