Havuz

Open-source agent sandbox infrastructure for parallel coding agents: isolated workspaces, MCP/CLI control, observability, and atomic publica

An open taxonomy and scoring framework for evaluating AI agent sandboxes: 7 defense layers, 7 threat categories, 3 evaluation dimensions, 27

One TypeScript API for running code in sandboxes. Write it once, run it on E2B, Daytona, Vercel Sandbox, Upstash Box, Railway, Tenki, Freest

self-extending sandboxed agentic harness. a loop, two tools and an llm

自托管 AI Coding Agent（Rust）：常驻 Runtime 撑起终端 / 浏览器 / 手机三端；每行改动可追溯到具体 Agent 与工具调用，子 Agent 的活由退出码裁决，检查点回退、目标 token 预算与验收清单跨轮次跨重启累计，用量与反馈账本留在本地。接任

Nosis is a lightweight personal agent runtime: per-session execution boundaries with OS-level sandboxing, computer use, a full built-in tool

Policy guardrails for coding agents (Claude Code, Codex, Cursor) — every tool call is checked locally, before it runs.

Open-source AI app builder — turn a prompt into a full-stack Next.js web app with built-in hosting, sandbox, database, auth, AI integration,

TypeScript interpreter for AI agents. Written in Rust. 2µs cold start. Sandboxed. Alternative to MCP tool calling.

OpenHermit is the open-source platform for deploying fleets of AI agents as production services — durable state, sandboxed execution, manage

An open-source workspace for coding agents, running on your own machine. More work. Less AI waste. Same subscriptions.

A framework for building intelligent agents

Fast cold-boot MicroVM sandboxes for AI agents on cocoon

Strong agent. Short leash

A hypervisor and sandbox cloud for self-hosting AI agents and RL

Sandbox for AI coding agents (Claude Code, opencode, pi): runs each project in an isolated microVM with host-side secret injection. Linux/KV

Open-source sandboxes for AI agents — run untrusted, AI-generated code safely on your own machine.

Aggregated Kubernetes apiserver and e2b-compatible data plane for warm-pooled agent sandboxes on microVMs. Sandboxes are served from per-nod

Runtime-neutral AI execution bridge for Claude Agent SDK, Codex app-server, and Agent Client Protocol (ACP).

A sandbox for the third-party APIs you depend on. Built from the provider's spec, it fails on purpose, and it replays the exact failure prod

Lightweight, secure Linux sandboxes for untrusted processes. Runs in the browser and on the server.

Sandbox for running AI coding agents autonomously: isolated, network-restricted, host-safe

A local control plane for AI agents — see what they do, approve what matters, keep secrets out. Rust + Tauri + Chrome MV3.

Hermes Agent (NousResearch) sandboxed by NVIDIA OpenShell — hardware-enforced network/filesystem/syscall policy, full memory + gateway stack

Run coding agents unattended. Agents write the code. Axocoatl writes the receipt. Claude Code, Codex or a local model in a sandbox; the host

Effect building blocks for agentic ai

Dispatch your task backlog to local, interactive AI coding agents. One git worktree per task, sandboxed by default.

Security-oriented three-axis observability for sandboxed AI agents: model intent, app context, and runtime telemetry into verifiable evidenc

Self-hosted execution platform for AI agents, powered by gVisor.

Manage OpenClaw in your team (Enterprise) by providing it compute infrastructure, tool integration, Authentication and security primitives

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

Durable Cloudflare computers for agents: run repos, steer live workspaces, verify results, and return receipts.

Capka AI — An open-source, self-hosted sandbox for AI agents. A community-driven alternative to Claude's Cowork

Finds and fixes agent network access risks in OpenShell and Docker sandboxes.

Local AI agent sandbox runtime with Firecracker, gVisor, and NVIDIA GPU sandbox support.

Sandboxed OCI container images for AI coding agents, built reproducibly with Nix.

Agent dashboard runs reliability scans, skills, and network access probes.

Open-source sandboxes for AI agents and code execution.

CoDA-Bench is a benchmark for code agents on data-intensive tasks. 🎈代码智能体能搞定数据密集型任务吗?

Verify what AI coding agents can actually read, write, execute and reach using synthetic canaries.

Docker sandbox for the Claude Code CLI. Runs Claude against one project folder — including with --dangerously-skip-permissions — while your

Micro-VM sandbox for AI agents: isolated, gated and auditable, to prove what your agents did

WorkBuddy Multi-Account Manager & Automated Check-in Toolkit (多账号轮换、全域工作区打通与自动化签到)

A desktop app for creating and managing isolated Mac environments.

Control plane for swarms of sandboxed AI agents: run, fork, pause, steer, and checkpoint agents that each work inside their own isolated san

Deterministic bytecode VM for agent execution - verified images, snapshots and a hash-chained replay journal

A cross-platform agent framework and samples, support on Windows/MacOS/Android/iOS/HarmonyOS

Confine an AI coding agent to one directory with a BPF-LSM program, and watch every escape it tries.

A block-based 3D space crafting game: explore procedurally generated planets across many star systems, mine and craft, tame creatures, build

Safe runtime for Claude Code, built to be yours.