Nicheloom

Market intelligence for builders — see what's gaining traction before it's crowded.

SmolVM

open-source sandbox for coding and computer-use agents

Details

External ID
47711887
Source
HN
Company
—
Product
SmolVM
Website domain
github.com
Launched
April 10, 2026
Cohort
—
Upvotes
8
Upvotes percentile
0.4813624678663239
Tags
—
Fetched at
Sept. 7, 2026, 9:26 p.m.
Updated at
Sept. 7, 2026, 9:26 p.m.

Description

SmolVM is an open-source local sandbox for AI agents on macOS and Linux.I started building it because agent workflows need more than isolated code execution. They need a reusable environment: write files in one step, come back later, snapshot state, pause/resume, and increasingly interact with browsers or full desktop environments.Right now SmolVM is a Python SDK and CLI focused on local developer experience.Current features include: - local sandbox environments - macOS and Linux support - snapshotting - pause/resume - persistent environments across turnsInstall: ``` curl -sSL https://celesto.ai/install.sh | bash smolvm ```I’d love feedback from people building coding agents or computer-use agents. Interested in what feels missing, what feels clunky, and what you’d expect from a sandbox like this.

Enrichment

Theme
developer tools for AI agents
Vertical
Horizontal
Function
Model & infra
Audience
Developer
AI stance
AI-native
Project type
Hobby / open-source project
Normalized one-liner
sandbox environment for ai agents
Manually corrected
False

Could you build this?

No Building a lightweight VM/sandbox with snapshotting, pause/resume, and browser automation on macOS and Linux requires low-level virtualization and hypervisor expertise.

What it would actually take: SmolVM relies on OS-level virtualization hypervisors like Apple's Hypervisor.framework / Virtualization.framework on macOS and KVM / Firecracker / QEMU on Linux, written in Rust or Go. The challenging parts are orchestrating memory snapshotting, copy-on-write root filesystems, guest-to-host vsock communication, and headless browser bridge integration. It demands deep systems engineering expertise in OS virtualization, kernel features, and hypervisors.

Discussion

3 comments analyzed.

Competitors mentioned: nono

Concerns raised: Whether sub-agents spawned in VM get killed by OOM if resource requirements exceed VM allocation, How gracefully OOM situations are handled

Feature requests: Elastic RAM configuration that expands based on usage, Adjustable resource limits for specific use cases

Competitors

Other products that read as similar to this one — 102 launches clear the similarity bar, closest 8 shown.

Attention rank: #51 of 103 (itself plus its competitors, highest first — normalized so YC and Product Hunt are compared fairly).

Launched 134 days after the earliest competitor.

Other launches for this product

Same idea, different domain

Nobody's really built a model & infra tool for Fintech yet.