Hi HN!<p>My name is Abhishek. I've spent my career working on Operating Systems and Infrastructure at places like Replit, Google, and Microsoft.<p>Today, I'm excited to launch Arrakis: an open-source and self-hostable sandboxing service designed to let AI Agents execute code and operate a GUI securely.<p>GitHub: <a href="https://github.com/abshkbh/arrakis" rel="nofollow">https://github.com/abshkbh/arrakis</a><p>Watch Claude build a live Google Docs clone using Arrakis via MCP – with no re-prompting or interruption.
<a href="https://www.youtube.com/watch?v=IZ5cAnhAdPQ" rel="nofollow">https://www.youtube.com/watch?v=IZ5cAnhAdPQ</a><p>Key Features<p>- Self-hostable: Run it on your own infra or Linux server.<p>- Secure by Design: Uses MicroVMs for strong isolation between sandbox instances.<p>- Snapshotting & Backtracking: First-class support allows AI agents to snapshot a running sandbox (including GUI state!) and revert if something goes wrong.<p>- Ready to Integrate: Comes with a Python SDK (<a href="https://pypi.org/project/py-arrakis/" rel="nofollow">https://pypi.org/project/py-arrakis/</a>) and an MCP server (<a href="https://github.com/abshkbh/arrakis-mcp-server" rel="nofollow">https://github.com/abshkbh/arrakis-mcp-server</a>) out of the box.<p>- Customizable: Docker-based tooling makes it easy to tailor sandboxes to your needs.<p>Sandboxes = Smarter Agents<p>As the demo shows, AI agents become incredibly capable when given access to a full Linux VM environment. They can debug problems independently and produce working results with minimal human intervention.<p>I'm the solo founder and developer behind Arrakis. I'd love to hear your thoughts, answer any questions, or discuss how you might use this in your projects!<p>Get in touch<p>- Email: abshkbh AT gmail DOT com
- LinkedIn: <a href="https://www.linkedin.com/in/abshkbh/" rel="nofollow">https://www.linkedin.com/in/abshkbh/</a><p>Looking forward to your feedback!