Capability-based sandboxes with fine-grained policies . Brokering access directly within the agent's operating context, with zero setup and zero latency

A Unified Virtual Filesystem For AI Agents

Nuwax Agent OS - The world's first universal agent operating system, building your private vertical general-purpose agent. 通用智能体操作系统，打造你私有的垂类通用智能体。新一代AI应用设计、开发、实践平台，无需代码，轻松创建，适合各类人群，支持多种端发布及API，提供完善的工作流、插件以及应用开发能力，RAG知识库与数据表存储能力，MCP接入以及开放能力。

Secure local dev environment for collaboration with AI coding agents

Agent-Sandbox is an E2B compatible easy-to-use enterprise-grade sandboxes for AI Agents. Allows Agents to securely run untrusted LLM-generated Code, Browser use, Computer use, and deploy Website etc.

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

SOCKS5 and DNS proxy for greywall

Sandboxed containers for AI coding agents with disposable Git/Jujutsu workspaces

Claude Code agent-in-container orchestration and automation

Lightweight Linux sandbox for AI agents. Kernel-native isolation (namespaces, cgroups, seccomp, Landlock) with REST API, MCP bridge, and web dashboard. Single Rust binary.

Claude-In-A-Box: Manage coding agent instances (Claude Code, Codex, Gemini CLI, Cursor) as local processes or containers. REST API, CLI, desktop app.

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

Open-source AI agent desktop sandbox. Give any AI a clean Linux+Chromium environment over HTTP — for testing, RPA, agents, scraping, demos.

Fast, Multi-Cloud Sandboxes for AI Agents.

Prompt-sanitizer runs entirely in-process: no cloud calls, no telemetry, no outbound dependency on third-party redaction APIs. In FAST mode it stays lean with zero ML dependencies. In SMART and FULL modes it adds fully local NER for names and organizations, plus bidirectional deanonymization, synthetic replacements, and audit logging.

Run AI coding agents (Codex, Claude, Gemini) in disposable Lima VMs and sync their work back through git, not rsync.

Enable lightweight Linux sandboxes with native kernel isolation for secure, scalable AI workloads using CLI and REST API access.

An agent sandbox with VFS and bash command-line interface for server-side AI agents. Built with .NET.

Secure multi-stage Dockerfile templates for production container images — distroless, non-root, hadolint-clean, with a tool-rich agent sandbox for LLM-generated code. Python first; Node, TypeScript, Go, Rust, JAX planned.

Wraps goose with bwrap to isolate things to the current directory along with some ENV