# gVisor

> Apache-2.0 userspace container sandbox — OCI runtime runsc for agent isolation

**Category:** DevOps  
**Pricing:** open-source  
**URL:** https://gvisor.dev  
**GitHub:** https://github.com/google/gvisor  
**Added:** 2026-10-06

---

## The Hook

gVisor is Google’s userspace application kernel for containers: the runsc OCI runtime sits between your workload and the host kernel so an agent sandbox (or any untrusted container) cannot treat “container escape” as a single kernel bug away.

## What's Good

Apache-2.0 (google/gvisor). Free path: install runsc and point Docker/Kubernetes/containerd at it — no account, no SaaS. Implements a Linux-like syscall surface in Go, integrates as a drop-in OCI runtime, and is the isolation layer many agent/sandbox products already sit on. Useful when you need stronger isolation than namespaces+seccomp without standing up a full VM per task. ★~19.6k.

## The Catch

Linux only (kernel 5.6+ recommended). Not every syscall is implemented — exotic workloads fail or fall back. Heavy file and network I/O pay a measurable overhead versus runc. This is a runtime, not an SDK or an orchestration product: you still wire Docker/K8s/agent harnesses yourself. Vs Firecracker/Cloud Hypervisor (microVMs) and plain seccomp: gVisor is the userspace kernel middle path.

## Verdict

Apache-2.0 OCI sandbox runtime when agent or untrusted containers need more than runc — Linux-only, syscall gaps, and you bring the orchestrator.