MCP server that keeps raw tool output out of AI coding agent context windows and persists session state across compaction.
Self-hosted AI agent platform with persistent bots, sandboxed computers, and bring-your-own model, available on web, Electron, and mobile.
Ephemeral Workspaces, durable shared history. LayerFS gives every agent a complete, writable filesystem Workspace over shared committed history, without copying the base. Agents work in parallel, commit what changed as deduplicated checkpoints, then branch, rewind or merge. Try it with LayerFS × DSH.
Desktop platform giving AI agents full macOS and Linux desktops in local VMs or your own cloud, with drivers, models, and benchmarks.
The ultimate open-source Overleaf extension and alternative: a free, self-hosted, drop-in replacement for Overleaf Community Edition with nearly all Pro feature—SSO, AI Assistant, Git(Hub/Lab)/ Dropbox sync, track changes, sandboxed compilation, Zotero, Mendeley and more. Formerly @Ayaka-notes Overleaf Pro. Docs: https://ayakaleaf-pro.ayaka.space
Zero-trust runtime that sandboxes and snapshots large numbers of untrusted agent workloads on standard Kubernetes.
OpenShell is a sandboxed runtime that enforces file, system call, and network policies on autonomous AI agents.
An open-source long-horizon SuperAgent harness that researches, codes, and creates. With the help of sandboxes, memories, tools, skill, subagents and message gateway, it handles different levels of tasks that could take minutes to hours.
Sandboxed Platform for Agentic Code Execution: pluggable-runtime sandboxes for running agent and user code in isolation.
A full-stack TypeScript CMS built on Astro, with sandboxed plugins and Portable Text content storage.
Terraria 1.4.5.8 Web Rewrite, using GLM-5.3-Flash and GLM-5.3
Omnigent is an open-source AI agent framework and meta-harness: orchestrate Claude Code, Codex, Cursor, Pi, and custom agents — swap harnesses without rewriting, enforce policies and sandboxing, and collaborate in real time from any device.
The World's First Virtual Terminal for AI Agents
AgenticX is a unified, production-ready multi-agent platform — Python SDK + CLI (agx) + Studio server + Machi desktop app. Features Meta-Agent orchestration, 15+ LLM providers, MCP Hub, hierarchical memory, avatar & group chat, skill ecosystem, safety sandbox, and IM gateway (Feishu/WeChat).
Secure, Fast, and Extensible Sandbox runtime for AI agents.
Low-level unprivileged sandboxing tool used by Flatpak and similar projects
Anti-Detect Sandbox: Chromium and Firefox fingerprint engines
agent-sandbox enables easy management of isolated, stateful, singleton workloads, ideal for use cases like AI agent runtimes and reinforcement learning (RL).
The first exam for code agents in the physical world: frozen robot tasks from many benchmark suites, each answered with a program and verified in a separate sandbox. Runs on Harbor.
Ultimate assistant for Sons Of The Forest. Features: God Mode, Infinite Health/Stamina, No Hunger/Thirst/Sleep, Perfect Body Temperature, Never Wet/Sick, Unlimited Ammo, Max Tree Damage, Infinite Durability, Infinite Flashlight Battery, Stealth Mode, Item Spawner, and more. 20+ functions.
Secure and fast microVMs for serverless computing.
Run agents like Hermes, LangChain Deep Agents, and OpenClaw more securely inside NVIDIA OpenShell with managed inference
agent runtime security - zero trust, zero setup, zero latency agent sandboxing
Creating realistic RL environments requires collaboration between researchers, engineers, and domain experts across many dimensions: artifacts, environments tools, dynamism of the environment, reproducibility, and more. There is no open source framework for building these environments effectively. Until now.
Composio powers 1000+ toolkits, tool search, context management, authentication, and a sandboxed workbench to help you build AI agents that turn intent into action.
✨ The agentic HTML editor — your local AI agent writes the HTML, you ship it. 🚀 75 Skills × 9 Surfaces (magazine · deck · poster · XHS / tweet · prototype · data report · Hyperframes) 🛡️ Sandboxed preview · 📤 1-click to WeChat / X / Zhihu / HTML / PNG 🔑 Zero API key — Claude Code / Cursor / Codex / Gemini / Copilot / OpenCode / Qwen / Aider.
Orchestrate sandboxed coding agents in TypeScript with sandcastle.run()
The Open Framework for Building Agents
A lightweight sandboxing tool for enforcing filesystem and network restrictions on arbitrary processes at the OS level, without requiring a container.
Sandboxed execution system that runs untrusted code on Windows, Linux, and macOS using OS-native sandboxes or VMs behind one policy model.
📚 《Deep Agents 实战》—— LangChain 官方大使出品,基于 LangChain / LangGraph 生态,从零构建生产级 AI Agent 的完整指南
Cross-Platform Micro-VM Sandbox for Agentic Workloads
The automation tower defense RTS
🔥 Open Source Browser API for AI Agents & Apps. Steel Browser is a batteries-included browser sandbox that lets you automate the web without worrying about infrastructure.
Bring pi coding sessions to Cua Spaces. Keep pi local, run workspace tools on your Linux and macOS computers, and sync Git changes back.
Context window optimization for AI coding agents. Sandboxes tool output (98% reduction), persists session memory via SQLite + FTS5, and enforces routing across 17 platforms through MCP + hooks. Works with Claude Code, Gemini CLI, VS Code Copilot, OpenCode, Codex CLI.
Silence iOS call-recording disclosure tones over USB — no app, no jailbreak, no signing, no VPN. macOS, one cable. Verified on iOS 27.0.1 (24A446).
Declarative Kubernetes-style orchestrator that runs sandboxed autonomous agent tasks on Agent Substrate, with workspaces and model specs.
Instant, Concurrent, Secure & Lightweight Sandbox for AI Agents.
an open-source alternative to the dots, bots and muses of the world, run an office of claude code/codex like agents on your laptop, sandboxes or anywhere, uses your existing subscriptions
Dory is the complete local development system for Apple Silicon: Docker, Compose, Kubernetes, virtual machines, and policy-bound agent sandboxes.
The SQLite of agent sandboxes — self-hosted, E2B-compatible. One machine, sandboxes that live forever, idle costs nothing.
A terminal coding agent built with DeepSeek and Windows in mind, featuring sandboxed commands, persistent sessions, and automated code verification.
Voxel sandbox game with a large render distance, procedurally generated content and some cool graphical effects.
Application Kernel for Containers
Open-source platform for creating safe, isolated production sandboxes for API, integration, and E2E testing.
Malware Configuration And Payload Extraction
Simple Android PDF viewer based on pdf.js and content providers. The app doesn't require any permissions. The PDF stream is fed into the sandboxed WebView without giving it access to content or files. CSP is used to enforce that the JavaScript and styling properties within the WebView are entirely static.
Open-source AI teammates in Slack, Teams, Discord, Telegram and Mattermost. Self-host it, with any agent harness and any model.
Self-hosted agent jobs that learn. Describe the task once, run it in a disposable sandbox and watch it turn into a script.
✨ AI Spreadsheet for Business
🚀 Fast and lightweight sandboxes for your apps and AI agents
A lightweight WebAssembly runtime that is fast, secure, and standards-compliant
Personal memory across agents
An embeddable, portable, branchable virtual machine to safely run Agents locally.
Chrome extension & CLI to let agents control your browser. Runs Playwright snippets in a stateful sandbox. Available as CLI or MCP
Open-source, self-hosted AI penetration testing framework: maps your attack surface into a graph, autonomously exploits it from a Kali sandbox with human approval gates, and opens PRs that fix what it finds. MCP both ways: plug in any MCP server as a tool, or drive RedAmon from Claude Code or your own agent.
1 place to call all your agents - OpenCode, Hermes, Claude Managed Agents, Cursor Agents API, DeepAgents.
Main development repository for Rigs of Rods soft-body physics simulator
The most compatible MCP CLI client. Supports: tools, resources, prompts, completions, async tasks, skills, stdio and streamable HTTP (MCP 2026-07-28 and 2025-11-25), persistent sessions, OAuth 2.1 with DCR, CIMD, client credentials, and ID-JAG, OS keychain, --json for code mode, grep for tool discovery, proxy for AI sandboxes, x402, and more.