Agent frameworks & infrastructure
Agent sandboxes & runtimes
1,721 agents · most starred first
Agent sandboxes & runtimes, page 12 of 18
Long-running process layer for CLEO — agent polling, SSE connections, heartbeat, key rotation
Daytona TypeScript SDK for sandbox management and code execution
Code Mode: use LLMs to generate executable code that performs tool calls
E2B SDK that give agents cloud environments
E2B Code Interpreter - Stateful code execution
Daytona TypeScript SDK for sandbox management and code execution
agent-swarm.dev — an open-source operating system for AI work: a lead agent breaks goals into tasks, routes them to specialized workers such as Claude Code or Codex, runs each worker in an isolated container, and preserves shared memory, tools, schedules,
Self-hosted AI agent framework for Convex + E2B with code execution
ROCK TypeScript SDK - Sandbox management and agent framework
OpenSandbox support for the Mastra AI agent framework
Agentic SDK for E2B's hosted code interpreter.
```js const pkg = require("agentic-os");
Library for managing agentic app-building containers
The runtime your AI agents deserve. Process management, memory bus, tool routing, message passing, and scheduling for autonomous agents.
Universal AI agent packaging format — bundle souls, memories, skills, tools, crons, subagents, and encrypted secrets into a single portable archive, then compile to any platform.
CLI installer for cage-bro sandbox — downloads the binary on install
CerebreX CLI — The open-source Agent Infrastructure OS
POSIX for AI agents — permissions, namespaces, resource quotas, content-addressed audit, downscoped handoff, server-side credential vaults, and rubric-graded outcomes. The OS-level substrate that runs above Modal-class sandboxes and below MCP/A2A.
Kubernetes for AI agents — manage fleets of AI workers
Standalone Agent Gateway for AionUI remote ACP sessions (OpenCode, Claude Code, Flowise, LangGraph)
LangGraph-based agent runtime for Hexos
Turn AI agents into microservices.
AgentBinary for MaaFw
I/O handler classes for SpiNNaker software stack
Docker-like CLI for AI agents with Enhanced LLM Intelligence
A barebone sandbox for agentic coding based on UNIX user
A distributed, scalable filesystem backend for deep AI agents, backed by blob storage
A CLI-driven virtual filesystem for AI agents
Agent Computer
Database Copy-On-Write for AI agent workspace isolation
Environmental Abstraction for General Agent.
Self-hosted file handoff server for AI agents — temporary upload and download links for agent↔user file exchange
Hyperlight CodeAct integrations for Microsoft Agent Framework.
Microsoft Agent Framework Runtime.
High-performance filesystem operations for AI agents
A toolkit for building, managing, and extending Jupyter agent workflows, including kernel and notebook utilities.
Safety-first sandbox execution for agents, CI, and MCP tools — Python SDK and optional embedded runner.
Local-first playground for experimenting with simple AI agents and sandboxed Python execution.
A simple SDK for deploying agent pods using FastAPI and SSE.
A private task queue system for agents using Redis + Celery with gRPC interface (Client Package)
Agent Realm — Multi-realm AI Agent deployment and management system.
Agent Runtime Manager Server
Python SDK for the All-in-One Sandbox API, >=1.7.0
agent-sandbox E2B SDK patch — one-line switch to use Agent Sandbox via the E2B Python SDK
Agent Sandbox plugin for Harbor — run Terminal-Bench / SWE-bench / Harbor benchmarks on Agent Sandbox pools
Production-grade self-correcting AI agent platform with sandboxed execution
This a for iwt internal tools package
Filesystem-backed agent runtime with an inbound FastAPI service above the LLM layer
Simulation-to-reality bridge for AI agents with environment adapters
Sovereign and edge deployment toolkit for self-contained agent bundles
A fast framework for building Docker-wrapped AI Agents
Run coding-agent CLIs in an isolated profile and workspace.
Local-first, governed AI agent runtime for Python — embed it in your app, or run it as a CLI or ACP server. Permissions, MCP, memory and audit replay built in.
AgentBox Python SDK that give agents cloud environments
Agent wrapper for bubblewrap
Defense-in-depth proxy sandbox for AI agents
AgentCloudService SDK - Cloud hosting for autonomous AI agents
Tiny CLI for pushing a single Python Strands Agent file to Amazon Bedrock AgentCore Runtime.
AgentCube CLI - A developer tool for packaging, building, and deploying AI agents to AgentCube
Python SDK for AgentCube Code Interpreter
A desktop for AI agents
A desktop for AI agents
Lightweight Linux namespace sandbox with persistent shell and instant reset
Secure runtime framework for AI assistants.
Placeholder package for AgentEnclave CLI.
Placeholder package for AgentEnclave Core.
Placeholder package for AgentEnclave Runtime.
Agent Environments
Generator for AI Environments
Python SDK for AgentFense - A sandbox infrastructure service for AI Agents with fine-grained file permission control
Official Python SDK for the AgentFM peer-to-peer compute mesh.
Simple file management and serving for agents
AgentFS Python SDK - A filesystem and key-value store for AI agents
File sharing for AI agents - Upload folders, get a URL
Framework for building agentic environments with tool integration
High-performance execution layer for AI agents — filesystem, multi-language code analysis, process management
Infrastructure, container images, and reusable GitHub Actions for autonomous AI agents
Jupyter Kernel containing an agent that can access to other kernels
Daytona deployment backend for Agentix
Docker deployment backend for Agentix
E2B deployment backend for Agentix
Shell + file I/O primitives for Agentix sandboxes
Sandboxed rollouts you call like typed Python (import as `agentix`)
Agent Kernel - Unified AI Agents Runtime
Python SDK for agentkernel — run AI coding agents in secure, isolated microVMs
Python SDK for agentmb — local Chromium runtime daemon for AI agents
Layer 3: The Framework - A deterministic kernel for zero-violation governance in agentic AI systems with POSIX-style signals, VFS, and kernel/user space separation
Public Preview — The Secure Nervous System for Cloud-Native Agent Ecosystems - Identity, Trust, Reward, Governance
Public Preview — AgentMesh Runtime: Execution supervisor for multi-agent sessions with privilege rings, saga orchestration, and audit trails
A persistent Python REPL for AI agents, driven from the CLI
AI Agent Framework — automates tasks using AI agents in Docker containers
Production-Ready AI Agent Runtime - Automate everything with LLM-powered agents
AgentOS Runtime — AI Agent CLI for workplace automation
Multi-language code execution against a real directory for AI agents
Package AI agents into portable .agent files