superserve-ai/superserve

Sandbox infrastructure for AI Agents

What it solves

Superserve provides persistent and secure sandboxes for AI agents, allowing them to execute code or perform tasks in an isolated environment without compromising the host system's security.

How it works

The project uses Firecracker microVMs to power its sandboxes, ensuring high-level isolation. It provides a suite of tools for management, including a TypeScript and Python SDK for integration, a CLI, and a dashboard console for monitoring and managing sandboxes.

Who it’s for

Developers building AI agents that require a secure, persistent environment to run code or interact with a system.

Highlights

  • Firecracker-powered: Uses microVMs for strong security and isolation.
  • Persistent Sandboxes: Environments that maintain state across sessions.
  • Multi-language Support: Dedicated SDKs for both Python and TypeScript.
  • Management Tools: Includes a CLI and a web-based console dashboard.

Related

  • Project
    opensandbox-group/OpenSandbox

    OpenSandbox is an open‑source sandbox platform for AI agents and other code‑execution workloads. It offers multi‑language SDKs, a CLI, and a server that can run sandboxes on Docker locally or on Kubernetes at scale, with strong isolation (gVisor, Kata, Firecracker), network policies, and a credential vault. Use cases include coding agents, browser/desktop automation, RL training, and secure code‑execution services.

  • Project
    tensorlakeai/tensorlake

    A compute infrastructure platform that provides secure, high-performance MicroVM sandboxes and serverless orchestration for running AI agents and LLM-generated code.

  • Project
    superradcompany/microsandbox

    A cross‑platform runtime that launches hardware‑isolated micro‑VMs (microVMs) from any OCI container image. Designed for untrusted workloads—AI agents, user scripts, CI jobs, browser automation—its microVMs boot in < 100 ms, can be controlled via a CLI (`msb`) or SDKs in TypeScript, Rust, Python, Ruby, and Go, and guarantee that secrets never leak into the VM. It’s aimed at giving AI agents a safe, fast sandbox for code execution.

  • Project
    kvcache-ai/AgentENV

    A scalable platform for running AI agent environments using Firecracker microVMs, featuring rapid snapshot-based booting and high-density resource management.

  • Project
    CelestoAI/SmolVM

    A framework for providing AI agents with secure, disposable microVMs to safely run code, browse the web, and interact with local files.