Suggestions appear as you type. Use the up and down arrows to choose one and Enter to open it.

exec-sandbox

Free#22 of 26 in AI Agent Sandboxes

exec-sandbox: A free, self-hosted QEMU sandbox with strong isolation and developer APIs. Ranked #22 of 26 in AI Agent Sandboxes by our editors (7.0/10); pricing: Free plan; best for self-hosted programmable QEMU sandboxes.

7.0/10Editor score
exec-sandbox7.0 Visit exec-sandbox

At a glance

  • Editor score
    7.0 / 10
  • Pricing
    Free plan
  • Best for
    Self-hosted programmable QEMU sandboxes
  • Free plan
    Yes
  • Paid from
    Free
  • Facts checked
    16 Nov 2026
  • Where it wins

    • Hardware-isolated QEMU microVM per session
    • Python asyncio API and sbx CLI with multi-language runtimes
    • Encrypted persistence, package caching, and network allowlists
  • Where it doesn't

    • Deployment is self-hosted on Linux or macOS hosts
    • Outbound networking starts disabled and requires domain allowlists
    • Support is provided through community channels and documentation

Our verdict on exec-sandbox

exec-sandbox is an open-source Python asyncio library and command-line tool for running untrusted code inside a hardware-isolated QEMU microVM for each session. It targets developers building AI agents or other systems that need isolated execution. The project runs on users’ Linux or macOS hosts without a cloud account or API key, and supports Python, JavaScript/TypeScript, and shell runtimes. Its self-hosted deployment model suits solo developers and small to mid-market teams that want control over the execution environment.

There is no paid tier: exec-sandbox is free and open source. The API and CLI expose session state, stdout and stderr streaming, file read/write, and versioned package installation with caching. Saved VM instances can be encrypted and stored locally or in S3-compatible storage. Network access is disabled by default, with configurable domain allowlists for controlled outbound connections. KVM, HVF, or TCG acceleration options let the runtime use available host capabilities, while the Python API supports asynchronous integration into agent workflows.

The trade-off is operational ownership. Because deployment is self-hosted, teams must run the Linux or macOS environment and manage the underlying execution setup rather than use a hosted sandbox service. Community support and documentation are the listed support channels, which may not fit organizations seeking vendor-managed assistance. Outbound access also requires deliberate allowlist configuration, and the repository documents execution timeouts up to 300 seconds. Choose exec-sandbox when programmable QEMU isolation, persistence, and local control matter; choose a hosted alternative when managed deployment or broader support coverage is a priority.

Your product? Our editors can test exec-sandbox in depth and publish a review, labelled as paid · from $149/mo →

exec-sandbox pricing

Plans Free planFree Free to use — no paid tier required for the core job.
See plans on github.com

exec-sandbox fact sheet

Free planYes
Paid fromFree
Browser sandboxNot verified
Network isolationYes
Persistent storageYes
Max runtimeNot verified
Runtime languagesMulti-language
Deployment modelSelf-hosted
GPU supportNot verified
DeploymentSelf-hosted
PlatformsLinux, macOS
SupportCommunity, Docs
Built forSolo, Small business, Mid-market (editorial estimate)
PricingFree plan
Websitegithub.com
Facts checked16 Nov 2026

Alternatives to exec-sandbox

See all exec-sandbox alternatives →

Used exec-sandbox? Be the first to review it

The editor score above is our own research. What this page doesn't have yet is a reader's view — what you used exec-sandbox for, what worked and what didn't. No stars are seeded and no review is paid for; an editor reads every one before it appears.

Write a reviewTwo minutes · verified accounts only · read by an editor before it appears

Reviews come only from verified accounts. Sign in or create an account first — your e-mail is never shown.

Your rating

0 characters · at least 80, up to 3,000

Posted from your verified account. Reviews appear after an editor reads them, usually within two working days.

Featured on iTechGuides

Featured on iTechGuides — exec-sandbox 7.0/10

exec-sandbox is listed in our AI Agent Sandboxes directory. Add the badge to your site — it links back to this page.

<a href="https://www.itechguides.com/products/exec-sandbox/"><img src="https://www.itechguides.com/best/badge/exec-sandbox.svg" alt="Featured on iTechGuides" width="230" height="46"></a>

Reviewed by iTechGuides Editors · Editorial team · Updated Nov 2026

Advertiser disclosure: iTechGuides is reader-supported. Vendors can pay for top positions in our rankings and for a place on other products' pages, and we may earn a commission when you click some links. Bought places stand at the number bought; this page and our disclosure say so. Paying vendors may get a fuller, more detailed review from our editors, and that review can change their score. How we rank.

Last updated · How we research and update