Agent frameworks & infrastructure

Agent sandboxes & runtimes

1,721 agents · most starred first

Agent sandboxes & runtimes, page 5 of 18

agent-images★ 50

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

fak★ 50

Agentic Runtime

grok-remote★ 50

Run grok agents on one machine, drive them from any device on your tailnet. Live web UI speaking ACP directly to grok agent stdio.

celln★ 50

Run agents in isolated cells that borrow verified tools instead of rebuilding Linux environments.

NovitaBox★ 50

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

pandastack-ai★ 50

Open-source sandboxes for AI agents and code execution.

scratch★ 50

Setup for Agent Computers

tetral★ 50

The open-source cloud agent platform

CodeFlow-harness★ 50

支持并发与多轮对话的 Agent Harness,具备上下文治理、断点恢复、Shell 沙箱和回归评测。

oz-agent-worker★ 50

Run Oz cloud agents on your own infrastructure with full orchestration and observability.

agenticlinux★ 50

An agentic distro with OpenClaw, llmman, Docker Sandboxes and Docker Engine pre-installed

cloudbox★ 49

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

pocket-pi★ 49

An agent-native runtime for embedded devices, powered by PocketJS. Build installable, updatable JavaScript apps for people and agents.

env★ 49

An agentic environment abstraction for Agent Substrate

nix-hermes-agent★ 49

Nix package and NixOS module for Hermes Agent by Nous Research

mvm★ 49

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

hoody-sdk★ 49

Computers that are URLs. TypeScript SDK & CLI for the Hoody platform - Terminals, files, browsers, AI agents, databases, and more. Pure HTTP.

bashbox★ 49

A sandboxed bash interpreter for AI agents, written in PHP

pi-on-cf★ 49

Run Pi Agent on Cloudflare

terrarium★ 49

Portable minimal MicroVM sandbox for agents and development tools.

picoaide★ 48

企业级AI PaaS管理平台 — 基于 Docker 的容器生命周期管理、LDAP/OIDC 统一认证、MCP 代理架构。让每个员工都有自己的 AI 代理。

gpt-code-interpreter★ 48

An independent, customizable version of OpenAI's code interpreter supporting multiple languages, unrestricted file access, and additional dependencies, executed safely within a Docker container. Contributions for UI improvements are welcomed!

addt★ 48

Run AI coding agents safely in Docker containers. Your code stays isolated - no surprises on your host machine.

arc★ 48

ARC Relay — WebSocket relay server for agent remote control by Axolotl AI

hermes-agent-docker★ 48

Simple docker sandbox image for Hermes Agent

gbash★ 48

Pure Go bash implementation for agent sandboxes

terok★ 48

Manager for podman containers for AI coding agents

oh-my-agentic-coder★ 48

oh-my-agentic-coder is a harness agnostic sandbox built on seatbelt/bubblewrap/landlock.

shipyard★ 47

⛵️ Yet another lightweight agent sandbox runtime, support python interpreter, shell, file system components, etc. 🏖️

agentd★ 47

The StereOS agent management daemon.

dify-agentbox★ 47

A production-ready, all-in-one Docker image designed for AI agents and autonomous systems that need to execute code across multiple programming languages.

claude-pod★ 47

Docker sandbox for the Claude Code CLI. Runs Claude against one project folder — including with --dangerously-skip-permissions — while your home directory, SSH keys, and other projects stay invisible to the container. Unofficial.

enter-os-core★ 47

The first operating system for AI agents — run heterogeneous agents as one governed, memory-driven, production-ready organization.

firetower★ 47

Run any coding agent, on your own servers, from anywhere

knecht-cloud★ 47

The home for the Knecht. Let agents boot, reproduce and test changes.

tunr★ 47

The network layer for AI agents. Reach your customers' on-prem data, show what you build, keep it running — through one relay that your machines only ever dial out to. Plus tunnels and deploys.

agenticlinux★ 47

An agentic distro with OpenClaw, llmman, Docker Sandboxes and Docker Engine pre-installed

agentos-railway-template★ 46

An agent platform you build, improve, and run using coding agents.

agenttier★ 46

Kubernetes-native sandbox platform to run AI agents, coding assistants and harnesses.

awesome-docker-sbx★ 46

A curated list of tools, kits, templates, integrations, and resources for Docker Sandboxes (sbx) running AI coding agents in isolated microVMs.

agent-os★ 46

The Open Agent Operation System

Clawee★ 46

Built by KrillinAI. Self-hosted enterprise Agent platform with local Codex Runtime, Desktop/Web workspace, MCP governance, RBAC, audit, Skill Hub, knowledge bases, and shared files.由 KrillinAI 打造。可自托管的企业级 Agent 平台,集成本地 Codex Runtime、桌面端与 Web 工作台、MCP 治理、RBAC、审计、Skill Hub、知识库和共享文件能力。

nemesis8★ 46

Container orchestration for agents.

awesome-AI-sandbox★ 46

A curated list of projects, tools, and references for running AI agents inside safer execution environments, with a focus on open-source solutions.

pomclaw★ 45

Enterprise AI Agent Platform: Distributed memory storage + SSH sandbox execution, serve unlimited agents with minimal cloud infrastructure

EmbedClaw★ 45

Decouple LLM, Tools, Agent, and Channels—then pack them onto a single ESP32

shai★ 45

sandboxing shell for ai coding agents

sanity-gravity★ 45

Providing a strong Gravity (sandbox) in the wild world of AI Agents (Antigravity, claude code, ...), to preserve the developer's Sanity.

jailoc★ 45

🔒 Jail OpenCode, Claude Code, Copilot CLI and OmO in micro-VMs: only your project folders, only the networks you allow.

ryu★ 45

Open-source agentic infrastructure & alternative to Grok Bot. Bring Claude Code, Codex, OpenClaw or your own. Start without building from scratch, add tools with plugins, and turn your agents into apps. Monitor every run and control what your agents can access.

box★ 45

TypeScript SDK and CLI for Upstash Box — sandboxed AI coding agents

autonomous-onboarding★ 44

Permissionless infrastructure for agents. No human intervention required.

Exploring-the-Code-Interpreter-in-OpenAI-GPT★ 44

The code interpreter is a tool developed by OpenAI to execute programming code in an interactive environment. It is capable of running Python code and displaying the results in real-time.

claude-agent-sdk-in-cloudflare-containers★ 44

A web application that runs Claude Agents in Cloudflare Containers using the Claude Agents SDK.

agentserve★ 44

A framework for hosting and scaling AI agents.

sheena★ 44

A lightweight, in-process sandbox for safely executing the tool calls an AI agent requests

agentbox★ 44

AI-powered sandbox environment with ephemeral Docker sub-agents for compute, browser, and app deployment

cmdop-sdk★ 43

Programmatic access to your CMDOP fleet from Python and Node

agex★ 43

Library-friendly Agents

boxlite-mcp★ 43

Powered by BoxLite - embeddable sandbox with hardware-level isolation and no daemon. The SQLite of sandbox, coming soon as open source.

AthenaOS★ 43

AthenaOS is a next generation AI-native operating system managed by Swarms of AI Agents

open-box★ 43

A desktop app for creating and managing isolated Mac environments.

tart-guest-agent★ 43

Guest agent for Tart VMs

senv★ 43

Build and serve Lean 4-verified OpenEnv reinforcement learning environments in Rust with h5i-app.

agro★ 43

🏗️ AGRO — Agent Governance Runtime Orchestrator. A portable home for autonomous coding agents.

kodosumi★ 42

runtime environment to manage and execute agentic services at scale

virgo-base-agent★ 42

A lightweight, cross-platform agent written in C with an embedded Luvit runtime

boringos★ 42

BoringOS is the open-source operating system for AI agents - it ships with a shell portal plus installable modules, powered by agentic CLIs.

sqlite-agent★ 42

A SQLite extension that enables SQLite databases to run autonomous AI agents, using other SQLite AI extensions.

agentos-railway★ 42

AgentOS: Serve agents over API, MCP, and chat interfaces like Slack

Tracekit-box★ 42

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

openclawer★ 41

One key Openclaw deployment - Deploy and manage OpenClaw AI agents with fnOS

agentwall★ 41

Run AI agents safely on your local machine

vigilante★ 41

Vigilante is a sandbox-first orchestration layer for coding agents. It isolates every task in a git worktree, enforces strict credential scoping, and gives you full audit logs — so your agents can't burn down production.

agent-circus★ 41

Run AI coding agents in sandboxed containers communicate via ACP

termo-agent★ 40

Open-source serverless runtime for AI agents. Bring any framework, get a production HTTP server with SSE streaming, sessions, auth, and memory.

my-ax★ 40

A personal AI agent operating environment you self-host on Cloudflare.

openLight★ 39

Lightweight AI agent runtime for homelabs, built around deterministic skills and local LLMs.

openclaw-kasmvnc★ 39

One-click deployment for OpenClaw + KasmVNC. Browser-based remote desktop with full container lifecycle management.

velane★ 39

Velane is agentic cloud for deploying your basic workflows, agents and sub-agents. 800+ OAuth integrations, sandboxed Bun and Python execution, and a full deployment pipeline managed via MCP

sandbox-runtime★ 39

A small, multi-architecture OpenSandbox runtime for production AI agents.

specter★ 38

Deploy AI agents to dedicated VMs in 90 seconds. Interactive TUI. Automatic DNS and TLS. You own the infrastructure.

polos★ 38

The open-source runtime for AI agents. Sandboxed execution with built-in tools, human-in-the-loop approvals, Slack integration, and durable workflows with automatic retries and prompt caching. You write the agent. Polos handles the infrastructure.

opencode-docker★ 38

A security-hardened Docker environment for running the OpenCode CLI in complete isolation.

morphz★ 38

Open-source agent runtime for persistent cognition, structured context, and concurrent work.

darwin-agentic-cloud★ 37

Verifiable and free cloud compute for AI agents. webMCP + MCP native. Check out our sandboxed Beta + research in the README

velane★ 37

Velane is agentic cloud for deploying your basic workflows, agents and sub-agents. 800+ OAuth integrations, sandboxed Bun and Python execution, and a full deployment pipeline managed via MCP

openharness★ 37

🏗️ Run coding agents in a sandbox, not on your machine.

autonomous-harness★ 37

One device to harness all AI agents — Claude Code, Codex, Cursor, OpenCode, and more — across computers and servers.

sprites-js★ 37

JavaScript and TypeScript SDK for Fly.io Sprites: computers for agents. Node-style APIs for creating Sprites, running remote commands, moving files, and managing checkpoints.

OpenDots★ 37

An open-source agent runtime inspired by OpenAI’s Dots: persistent AI agents that pursue goals, react to events, and keep work moving 24×7.

reasonix-portakit★ 36

Reasonix 便携工具箱:PortaKit for 0.5X(路径重定向)+ PortaKit for 1.X(NTFS Junction),U盘/同步盘即插即用。搭配 migration-assistant 实现 0.5X→1.X 无缝升级。 / Reasonix portable toolkit: PortaKit for 0.5X (path redirection) + PortaKit for 1.X (NTFS Junction), USB/sync disk plug-and-play. Pair with migration-assistant for seamless 0.5X→1.X upgrade.

autopg★ 35

Embedded PostgreSQL 18 server with true concurrent connections - zero config, auto-provision databases

ash-ai★ 35

Deploy Claude agents as production APIs — with sessions, streaming, sandboxing, and persistence handled for you.

RunDiffusion-Agents★ 35

Your Agent Farm control plane for OpenClaw, Codex, Claude, Gemini, Hermes, and the recovery tools that keep them healthy.

agent-sandbox★ 35

A sandboxed execution environment for AI agents via WASM

EnClaws★ 35

企业级的数字员工容器化管理平台

aisi-sandboxing★ 35

The open-source AISI toolkit for sandboxing agentic evaluations

agent-sandbox★ 35

Give any AI agent its own disposable Linux machine. Firecracker microVM sandboxing with millisecond boot times, full network access, and native MCP support.

apx★ 35

Meet your crew. Your own team of AI agents, open source and on your machine: a super-agent that answers you on WhatsApp, Telegram or the desktop, and a crew per project that works with Claude Code and Codex.

More in Agent frameworks & infrastructure