#sandbox

Dispatch your task backlog to local, interactive AI coding agents. One git worktree per task, sandboxed by default.

Open-source agent runtime — SSH-native isolation, eBPF egress policy, Kubernetes + LXC backends, GPU passthrough, MCP-native CLI

Chat, specs, tasks, and code. An autonomous engineering platform. Full autonomy when you trust it. Full control when you don't.

A secure, self-improving agent operating system in a single Go binary. Bring any model, manage local models, point it at a goal, and grant it real authority: every action is sandboxed, governed, and sealed into a verifiable, tamper-evident record an independent party can check. Runs interactive or 24/7, or embed it in your own system.

Toolkit and libraries for running AI agents inside microVMs

CLI tool for running coding agents inside hardware-isolated microVMs

A modern solution for running untrusted code in Node.js

AI agent sandboxing done right.

Millisecond microVM sandbox forking for AI agents on Kubernetes. Firecracker VMs that restore from memory snapshots in milliseconds, fork a running VM into N copies, and persist durable, versioned workspaces. Self-hostable, declarative CRDs.

Lightweight MicroVM engine built on Cloud Hypervisor. Features include OCI and cloud image support, instant snapshot and clone via reflink, Windows 11 guest support, CNI networking with TC redirect, memory balloon, hugepages, and a Docker-like CLI. Designed for AI sandboxing, cloud desktops, and ephemeral dev environments.

Open-source, self-hosted AI app builder — an agent builds real apps in isolated sandboxes on your own server, each live at a preview URL. Self-host in one command. MIT.

AI partners for every person: multi-user, multi-agent memory, trusted tools, schedules, sandboxed workspaces, and chat.

A Claude Skill to give your agent the ability to use a web browser

Embeddable Wasm sandbox for running untrusted mruby code from Ruby applications.