Running Claude Code dangerously (safely)

kept by eddie jan 21

Claude Code users discuss practical sandboxing strategies—VMs, containers, and firewalls—to safely run the agent with dangerous file-system permissions.

I’ve built a virtual museum with nearly every operating system you can think of
A virtual museum ships a Linux VM pre-loaded with 1700+ installs across 570+ distinct operating systems, all pre-configured and ready to run under emulation.
also on VirtualBox, #developer-tools
GitHub - apple/container: A tool for creating and running Linux containers using lightweight virtual machines on a Mac. It is written in Swift, and optimized for Apple silicon.
Apple released an open-source tool, `container`, that creates and runs OCI-compatible Linux containers as lightweight VMs on macOS 26+ Apple Silicon Macs.
also on #virtualization
Docker Sandboxes | Sandboxes for Coding Agents
Docker Sandboxes provides disposable microVM isolation for AI coding agents like Claude Code and Codex, enabling safe, unattended execution without host impact.
also on Docker, #security, Claude Code, #ai-coding, #developer-tools
Development environments for your cloud agents
Cursor introduces new tools for configuring cloud agent development environments, including multi-repo support, Dockerfile-based configuration, and enhanced security controls.
also on Docker, #security, #ai-coding, #developer-tools