code-sandbox

Give the agent a machine. Just not yours. Each AI coding agent gets its own isolated machine with root, Docker, and systemd - active defense detects and stops threats automatically.

OpenMA handles sessions, execution and recovery, so you can build and run agents with any model, any harness, any sandbox, on any cloud. An open-source, self-hostable alternative to Claude Managed Agents and OpenAI Agents API.

Open-source agent runtime — SSH-native isolation, eBPF egress policy, Kubernetes + LXC backends, GPU passthrough, MCP-native CLI

Autonomous background coding agents on AWS. Turn tasks into pull requests via isolated runtimes, with built-in orchestration, observability, and governance.

Sandboxes for running code and tools

Unified execution environment for Python code, shell commands, and programmatic MCP tool calls.

An interactive offline desktop application for visualizing computer science algorithms, built in JavaFX, featuring time-travel step controls (forward and backward) and a fully featured local code compilation sandbox for C, C++, Java, and Python.

Manage secure microVMs — one command from zero to secure sessions, locally and remotely.

Paid APIs and supported tokenized shares for agents, funded with prepaid balance. Instinct Checkout guidance: https://vaaya.ai/llms.txt

Stealth desktop AI assistant & interactive web showcase for CorvusX built with React 19, TypeScript, and Vite. Features a 60 FPS 3D particle canvas engine and a simulated Windows desktop environment with a sandboxed JS execution engine.

A minimal JavaScript playground/codesandbox for quick code testing

基于 Docker 的在线代码执行沙箱系统，支持 Java、JavaScript、Python、C 等多语言代码的安全隔离运行。采用 Spring Boot + Docker 技术栈，应用策略模式和模板方法模式实现代码的编译执行和资源限制。可作为在线评测系统的核心组件，是一个适合作为毕业设计参考或个人练手的项目，同时很适合放在简历上提高竞争力。

Sandboxed code execution daemon with enforced memory, CPU, and output limits. JSON API for humans and AI agents.

Self-hosted AI workstation for a compatible Linux VPS: agent chat, project workspaces, apps, shared browser, remote desktop, terminal and optional mail. MIT licensed.

GenPark AI Agent Skill - In-process isolated Python namespace REPL executor with stdout/stderr capture, safe builtins restrictor, and runtime variable introspection.

GenPark AI Agent Skill - In-process isolated Python namespace REPL executor with stdout/stderr capture, safe builtins restrictor, and runtime variable introspection.

Open-source online judge evaluation cloud with Java, Spring Boot, microservices, RabbitMQ, Docker Compose, and a code sandbox workflow.

A digital analogue clock

一个判题沙箱；可根据判题数据，判定用户程序的运行结果以及获取用户程序运行时间和内存消耗、可隔离程序网络。

An MCP server that provides a secure isolated environment for executing Python code.