arrakis
A fully customizable and self-hosted sandboxing solution for AI agent code execution and computer use. It features out-of-the-box support for backtracking, a simple REST API and Python SDK, automatic port forwarding, and secure MicroVM isolation. Perfect for safely running, testing, and backtracking
Arrakis is a self-hosted sandboxing solution with an MCP server that allows AI agents to safely execute code and perform computer use tasks in isolated MicroVM environments. It provides secure isolation for running untrusted agent-generated code, automatic port forwarding to access sandbox GUIs, and snapshot-and-restore capabilities that enable agents to backtrack to previous states. The MCP server integrates with Claude Desktop, Windsurf, Cursor, and other MCP clients, exposing REST API functionality through the Model Context Protocol. This makes it particularly valuable for agents executing multi-step plans that benefit from checkpointing and rollback capabilities.
Key Features
Use Cases
- 01Safely executing untrusted code generated by AI agents without risking the host system
- 02Implementing Monte Carlo Tree Search agents that explore multiple execution paths via snapshots
- 03Running computer use tasks where agents interact with graphical applications through VNC
- 04Testing and debugging agent-generated code in isolated, reproducible environments
- 05Creating checkpoints during complex multi-step agent workflows for explainability and debugging
- 06Building agents that need to rollback and retry failed execution paths
Related MCP Servers
View moremarkitdown
Python tool for converting files and office documents to Markdown.
firecrawl
The context API to search, scrape, and interact with the web at scale. 🔥
prompts.chat
f.k.a. Awesome ChatGPT Prompts. Share, discover, and collect prompts from the community. Free and open source — self-host for your organization with complete privacy.
langflow
Langflow is a powerful tool for building and deploying AI-powered agents and workflows.
arrakis — FAQ
What is Arrakis MCP server?+
Arrakis is an MCP server that provides AI agents with secure, isolated sandboxes for code execution and computer use. It runs each sandbox in a MicroVM and supports snapshot-and-restore for backtracking to previous states.
How do I install the Arrakis MCP server?+
First, set up arrakis-restserver on a Linux machine with KVM virtualization support using the setup script. Then configure your MCP client (like Claude Desktop) to connect to the arrakis-mcp-server repository with the appropriate command and arguments in the configuration file.
Which AI clients work with Arrakis?+
Arrakis MCP server works with any MCP-compatible client including Claude Desktop, Windsurf, and Cursor. The Python SDK can also be used programmatically with any AI agent framework.
What are the prerequisites for running Arrakis?+
You need a Linux machine with KVM virtualization enabled (/dev/kvm available). The README provides specific GCP setup instructions for running on Google Cloud Engine VMs.
Is Arrakis free to use?+
Arrakis is licensed under GNU AGPL v3.0 for open source use. For commercial licensing, you need to contact the author at abshkbh@gmail.com.
How does snapshot-and-restore work?+
Agents can create snapshots of a running sandbox at any point, preserving all processes, files, and state. Later, they can restore from that snapshot to return the sandbox to that exact state, enabling backtracking in multi-step plans.
How do I install arrakis?+
Open the source repository on GitHub and follow its README. arrakis is a mcp server — MCP Agents Market links you directly to the official repo.
Is arrakis free?+
arrakis is an open-source project hosted on GitHub. Check the repository for its license and any usage requirements.