About Agent Sandboxes
Agent Sandboxes (agentsandboxes.org) is an independent reference for developers choosing where their AI agents run code. We document 21 active sandbox providers and open-source projects, compare them side by side, and link every factual claim to the provider’s own documentation.
What this site is
AI agents increasingly write and run code, operate browsers and work on repositories for hours at a time. Choosing the sandbox they run in is now an infrastructure decision with real security and cost consequences — and the market has grown from a handful of providers to dozens in two years.
This site exists to make that decision faster and better informed: a directory of providers, a comparison table, head-to-head pages, use-case shortlists, and guides that explain the underlying technology.
Independence
- No provider pays for inclusion, placement or ranking.
- If we ever accept sponsorship, it will be clearly labelled “Sponsored” and kept separate from rankings and recommendations.
- Facts come from providers’ public documentation, pricing pages and repositories, and are dated. Editorial judgments are labelled as such and explained on the methodology page.
Contact and corrections
Providers and readers can submit corrections or new listings. Email: hello@agentsandboxes.org.
Cite this page
Agent Sandboxes. (2026). About Agent Sandboxes. agentsandboxes.org. Last reviewed November 8, 2026. https://agentsandboxes.org/about/
Found an error or a missing sandbox? Submit a correction — every entry links to the sources it relies on.