shedix

Blog

Posts about security

  1. Seven coding-agent disasters, and which ones a separate machine would have stopped

    Seven real incidents where coding agents deleted files, disks or databases, where each agent ran, and which ones a separate machine would have stopped.

    7 min read · security, agents, claude code, git

  2. Container, microVM or full VM: choosing a box for a coding agent

    What containers, gVisor, microVMs and full VMs each put between a coding agent and your computer, when the agent needs Docker, and what none of them fix.

    6 min read · security, agents, claude code

  3. Home server, VPS or cloud VM: where to keep your coding agent

    A computer at home, a monthly VPS or an hourly cloud VM for your coding agent: the cost at 4 hours a day, 8 hours a day and always on, and what can break.

    6 min read · agents, remote dev, costs, security

  4. VS Code on a remote machine: Remote-SSH, code-server or a terminal

    Three ways to edit code on a remote machine, what each one installs and where its extensions run, and which keeps your laptop out of an agent's reach.

    5 min read · remote dev, ssh, security, agents

  5. Why every shedix machine is a whole VM, not a pod

    Why shedix gives each machine its own VM instead of a container on a shared cluster, what that gives an agent, and what it costs in boot time and density.

    6 min read · agents, remote dev, security

  6. Give your coding agent its own machine

    What a coding agent can reach on your main computer, what Claude Code and Codex permission modes and sandboxes cover, and why a separate VM is cleaner.

    5 min read · agents, security, claude code, codex