BVisor – An Embedded Bash Sandbox, 2ms Boot, Written in Zig

Name: BVisor – An Embedded Bash Sandbox, 2ms Boot, Written in Zig
Availability: InStock
Author: edunteman

by edunteman·Feb 23, 2026·24 points·9 comments

Visit Project View on HN

AI Analysis

●●●BangerWizardryBig Brain

gVisor-inspired—2ms userspace sandbox beats containers for ephemeral agent tasks.

Strengths

•Seccomp user notifier + copy-on-write filesystem isolation achieves strong security without VM/container overhead—novel architecture for this use case
•2ms boot time is orders of magnitude faster than Docker/Kubernetes for agent workloads; native speed execution with minor syscall overhead
•TypeScript SDK + npm packaging lowers friction vs. pure syscall-level libraries; Python SDK planned

Weaknesses

•PoC status (early, Linux-only)—Windows/macOS support missing; documentation is sparse on threat model or production readiness
•No evidence of hardening against timing attacks or advanced seccomp bypasses—"should not be used in production" is honest but limits immediate impact

Post Description

bVisor is an SDK and runtime for safely executing bash commands directly on your host machine. We built it on the belief that "sandbox" doesn't need to mean shipping off to remote sandbox products, or spinning up local VMs / containers. Sometimes, you just want to run that bash command locally.

bVisor boots a sandbox from user-space without special permissions, powered by seccomp user notifier. This allows us to intercept syscalls from guest processes and selectively virtualize them to block privilege escalation, isolate process visibility, and keep filesystem changes isolated per sandbox (copy-on-write). Sandboxes boot in 2ms, and can run arbitrary binaries at native speed (with minor overhead per syscall). This approach is heavily inspired by Google's gVisor.

As of today, bVisor supports most filesystem operations, basic file I/O, and can run complex binaries such as python interpreters. It is packaged as a Typescript SDK and installable via npm. There's much to still implement (such as outbound network access to support 'curl', shipping a python SDK, etc), but we wanted to share it here for feedback and anyone who'd be able to make use of the current featureset!