Agent Code Sandboxing AI Agents

Tools and platforms for isolating, containerizing, and safely executing untrusted code from AI agents using Docker, VMs, WebSockets, or namespace-based sandboxes. Does NOT include general container orchestration, deployment platforms, or security monitoring without execution isolation capabilities.

There are 113 agent code sandboxing agents tracked. 3 score above 70 (verified tier). The highest-rated is e2b-dev/E2B at 92/100 with 11,263 stars and 3,818,023 monthly downloads. 9 of the top 10 are actively maintained.

Get all 113 projects as JSON

curl "https://pt-edge.onrender.com/api/v1/datasets/quality?domain=agents&subcategory=agent-code-sandboxing&limit=20"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.

#	Agent	Score	Tier	Stars	Language
1	e2b-dev/E2B Open-source, secure environment with real-world tools for enterprise-grade agents.	92	Verified	11,263	MDX
2	alibaba/OpenSandbox OpenSandbox is a general-purpose sandbox platform for AI applications,...	87	Verified	7,681	Python
3	e2b-dev/infra Infrastructure that's powering E2B Cloud.	76	Verified	960	Go
4	boxlite-ai/boxlite Sandboxes for every agent. Embeddable, stateful, snapshots, and hardware isolation.	64	Established	1,524	Rust
5	always-further/nono Secure, kernel-enforced sandbox CLI and SDKs for AI agents. Capability-based...	63	Established	980	Rust
6	eugene1g/agent-safehouse Sandbox your local AI agents so they can read/write only what they need	59	Established	1,184	Shell
7	zerobootdev/zeroboot Sub-millisecond VM sandboxes for AI agents via copy-on-write forking	57	Established	1,592	Rust
8	multikernel/sandlock Lightweight process-based sandbox for Linux, no container, no VM, no root.	57	Established	16	Python
9	agbcloud/agbcloud-sdk AI-native cross-platform sandboxes for developers, featuring multimodal...	54	Established	614	Python
10	adammiribyan/zeroboot Sub-millisecond VM sandboxes for AI agents via copy-on-write forking	54	Established	596	Rust
11	HACKE-RC/Bandsox Sanboxes for AI agents and humans	53	Established	70	Python
12	angelorc/vmsan Firecracker made simple. Spin up secure microVMs in milliseconds, from...	53	Established	28	TypeScript
13	dtormoen/tsk-tsk Keeping your agents out of trouble with sandboxed coding agent automation	51	Established	148	Rust
14	tomascupr/sandstorm Run Claude agents in secure cloud sandboxes — via API, CLI, or Slack. One...	50	Established	425	Python
15	mavdol/capsule A secure, durable runtime to sandbox AI agent tasks. Run untrusted code in...	47	Emerging	225	Rust
16	Th0rgal/sandboxed.sh Self-hosted orchestrator for AI autonomous agents. Run Claude Code & Open...	45	Emerging	300	Rust
17	runtm-ai/runtm Open-source sandboxes where coding agents build and deploy. Spin up isolated...	45	Emerging	38	Python
18	adarsh9780/safe-py-runner A lightweight, secure-by-default Python code runner designed for LLM agents.	43	Emerging	2	Python
19	projecteru2/cocoon Lightweight MicroVM VMM built on Cloud Hypervisor for AI	42	Emerging	33	Go
20	EXboys/skilllite A lightweight secure Self-evolution engine built in Rust, featuring a...	41	Emerging	48	Rust
21	railroad-dev/railroad Run Claude Code at full speed, safely. OS-level command blocking with no...	41	Emerging	17	Rust
22	rpfilomeno/opencode-docker Stop prompt injection catastrophe! Run your AI Agents in secure isolated...	39	Emerging	11	Shell
23	branchbox/branchbox Parallel, isolated dev environments for humans and AI coding agents. Real...	39	Emerging	63	Rust
24	Parassharmaa/agent-sandbox A sandboxed execution environment for AI agents via WASM	39	Emerging	13	Rust
25	onyx-dot-app/python-sandbox Secure and lightweight Python code execution environment for LLMs	38	Emerging	12	Python
26	legionus/devkit The project allows you to manage isolated containers with AI agents	37	Emerging	2	Makefile
27	89luca89/clampdown Run AI coding agents in hardened container sandboxes.	37	Emerging	32	Go
28	jamesmurdza/upstream-agents Run AI coding agents in isolated sandboxes connected to your GitHub repositories	37	Emerging	4	TypeScript
29	stacklok/brood-box CLI tool for running coding agents inside hardware-isolated microVMs	36	Emerging	19	Go
30	foundry-works/foundry-sandbox Ephemeral, batteries-included Docker workspaces that isolate AI coding...	36	Emerging	2	Python
31	rivet-dev/secure-exec Secure Node.js Execution Without a Sandbox A lightweight library for secure...	36	Emerging	10	TypeScript
32	joshualamerton/agentic-sandbox Simulation environment for testing and validating autonomous agents	35	Emerging	1	Python
33	gizmax/Sandcastle Production-ready AI agent workflow orchestrator. 63 integrations, EU AI Act...	35	Emerging	24	Python
34	getlark/runtimeuse Run AI agents inside sandboxes over WebSockets	34	Emerging	4	TypeScript
35	Raynan00/sandpy Browser-native Python sandbox for AI agents	33	Emerging	10	TypeScript
36	akshayaggarwal99/boxed The Sovereign Code Execution Engine for AI Agents. Run untrusted code safely...	33	Emerging	10	Go
37	the-void-ia/void-box Composable agent runtime with enforced isolation boundaries	33	Emerging	19	Rust
38	ixchio/agent-sandbox-runtime A secure runtime for self-correcting AI agents with Docker sandboxing.	32	Emerging	7	Python
39	STONE-CELL-SPF-JOSEPH-STONE/SPFsmartGATE AI Security Gateway — Compiled Rust enforcement between AI agents and your...	32	Emerging	7	Rust
40	deevus/pixels Disposable Linux containers for AI coding agents, with extensible backends	32	Emerging	50	Go
41	HappyHackingSpace/sindoq AI Sandbox	31	Emerging	10	Go
42	reoring/botbox Kubernetes sidecar that sandboxes container egress. Deny-by-default...	30	Emerging	12	Rust
43	Orellius/Laminae AI personality, safety, red-teaming, and sandboxing in Rust SDK.	30	Emerging	8	Rust
44	AxeForging/aigate OS-level sandbox for AI coding agents - kernel-enforced file, command, and...	27	Experimental	12	Go
45	arcboxlabs/arcbox Run AI agents on real and isolated machines — own kernel, filesystem, and...	27	Experimental	10	Rust
46	kajogo777/the-agent-sandbox-taxonomy An open taxonomy and scoring framework for evaluating AI agent sandboxes: 7...	27	Experimental	23	Go
47	seznam/jailoc 🔒 Jail your AI agents — sandboxed Docker environments with network isolation...	27	Experimental	9	Go
48	sevorix/sevorix-lite Sevorix Lite is a Rust-native, open-source runtime containment engine for...	27	Experimental	9	Rust
49	Parassharmaa/agent-fetch Sandboxed HTTP client with SSRF protection for AI agents. Prevents DNS...	26	Experimental	4	Rust
50	numcys/sudomode The Missing sudo Command for AI Agents.	26	Experimental	6	Python
51	hyperterse/sandboxer Single, consistent interface to run code, manage files, and control isolated...	26	Experimental	5	TypeScript
52	danievanzyl/pyro Open-source Firecracker microVM sandbox platform for AI agents	25	Experimental	3	Go
53	haasonsaas/capsule-run Lightweight, secure sandboxed command execution for AI agents	25	Experimental	2	Rust
54	dredozubov/hazmat macOS containment for AI agents — user isolation, kernel sandbox, pf...	24	Experimental	2	Go
55	opencapsule/opencapsule Secure Code Execution Runtime for AI Agents	24	Experimental	2	Rust
56	DavidKim0326/DUDA Isolation Guardian for Claude Code — Prevent AI agents from breaking...	24	Experimental	2	Python
57	qhkm/zeptocapsule Isolation sandbox for AI agents — process, namespace, and Firecracker capsules	24	Experimental	2	Rust
58	c4rb0nx1/tuprwre That install command your AI agent just ran? Never touched your host....	23	Experimental	5	Go
59	nhevers/agent-sandbox Sandboxed code execution for AI agents	23	Experimental	7	Python
60	liut/strata Lightweight Session Sandbox Service — Isolated Shell Environments via...	23	Experimental	1	Go
61	sauravbhattacharya001/ai Contract-enforced sandbox for studying AI agent self-replication safety	23	Experimental	1	Python
62	0rzech/vibe-containers Simple sandbox Podman containers for Mistral Vibe	23	Experimental	1	Shell
63	Mickdownunder/atlas-validation-layer Bounded validation and sandbox layer for the Operator research control plane	23	Experimental	1	Shell
64	Embedded-Focus/agent-circus Run AI coding agents in sandboxed containers communicate via ACP	23	Experimental	1	Python
65	KometzRobot/capsule-spec Open tools for AI identity persistence — Capsule Spec, Loop Harness, Cinder Enhanced	23	Experimental	1	Python
66	heromen22/sandstorm 🚀 Run multiple AI agents securely in isolated cloud sandboxes for long tasks...	22	Experimental	—	Python
67	nothingnesses/agent-images Sandboxed OCI container images for AI coding agents, built reproducibly with Nix.	22	Experimental	—	Nix
68	dklymentiev/screenbox Real virtual desktops for AI agents. MCP-native, self-hosted, fully isolated.	22	Experimental	—	Python
69	madeinplutofabio/command-scope-contract Protocol for bounded shell and CLI execution with explicit scope, policy,...	22	Experimental	—	Python
70	cyruscyliu/agent-vault Run AI coding agents in isolated Kata Container workspaces on k3s with tmux,...	22	Experimental	—	Python
71	SatishoBananamoto/svx Simulate, Verify, Execute — a safety layer for coding agents	22	Experimental	—	Python
72	edlsh/pi-extension-e2b E2B cloud sandbox integration for pi — redirects all tool execution to a...	22	Experimental	—	TypeScript
73	ClawWorksCo/lasso-sandbox LASSO — Layered Agent Sandbox Security Orchestrator. Sandboxed execution for...	22	Experimental	—	HTML
74	al002/agent-fort Security runtime for AI agents	22	Experimental	—	Rust
75	bird/paranoid Isolated QEMU microVM sandboxes with WireGuard-only networking for AI agents	22	Experimental	—	Shell
76	Daaboulex/openviking-nix OpenViking packaged for NixOS — agent-native context database for AI agents	22	Experimental	—	Nix
77	NihalKA/sandboxshift Self-hosted AI agent sandbox with automatic local/cloud bursting	22	Experimental	—	Python
78	DynamicExploit/runtm 🌐 Spin up isolated environments for coding agents to build and deploy...	22	Experimental	—	Python
79	Enigma-s9v/chitin-shell Protect AI agents by isolating LLMs from sensitive data with process...	22	Experimental	—	TypeScript
80	Rookie481/spotdb 🏖️ Create a secure, temporary data sandbox for AI workflows and exploration,...	22	Experimental	—	Go
81	ian-flores/securer Sandboxed R code execution with tool-call IPC for LLM agents	22	Experimental	—	R
82	rankgnar/agent-sandbox Linux-native sandboxing for AI coding agents. Run Codex, Claude Code, and...	22	Experimental	—	Shell
83	ammmir/sandboxer Forkable code execution server for LLMs, agents, and devs	22	Experimental	15	Python
84	Arjun2729/Ithilien Safe autonomous mode for AI coding agents. Docker sandbox + tamper-evident...	22	Experimental	—	TypeScript
85	Mykazi127/noxrunner 🔧 Interact with NoxRunner-compatible sandbox execution backends using this...	22	Experimental	—	Python
86	D8k4/clampdown Contain AI coding agents within secure container sandboxes that limit...	22	Experimental	—	Go
87	us/den Secure sandbox runtime for AI agents	21	Experimental	2	Go
88	agentbox-cloud/agentbox AgentBox SDK — Enterprise AI Sandbox Tools	21	Experimental	2	Python
89	throwparty/litterbox Review outputs, not actions: give your AI agents litter trays to poop into.	21	Experimental	2	Rust
90	nwcnwc/warden-proxy A localhost proxy that gives browser-sandboxed applications safe, controlled...	20	Experimental	1	Rust
91	lzjever/noxrunner Python client library and CLI for sandbox execution backends (NoxRunner...	20	Experimental	1	Python
92	YujiSuzuki/ai-sandbox-dkmcp Secure AI sandbox for Claude Code / Gemini — hide secrets, enable...	20	Experimental	1	Go
93	scarab-project/scarab-runtime Strict-confinement sandbox for autonomous AI agents. Built in Rust using...	20	Experimental	1	Rust
94	SudoDog-official/SudoDog Secure sandbox for AI agents. Blocks dangerous operations, monitors...	19	Experimental	—	Python
95	RutgerLubbers/cage Put untrusted commands in a cage. Flexible file system sandboxing with...	19	Experimental	—	Go
96	milyas2001/forge-agent-sandbox FORGE - Bare-Metal Microkernel for AI Agent Sandboxing	19	Experimental	—	Rust
97	geraldthewes/python-executor Currently the About section may be empty or generic. Suggested (107...	19	Experimental	—	Go
98	firmo-tecnologia/devbox Run unattended, safelly cloud code inside a container.	19	Experimental	—	Go
99	brooksomics/llm-rustyolo Secure Docker wrapper for AI coding agents with filesystem, privilege, and...	18	Experimental	7	Rust
100	danieljhkim/DevBox DevBox is a minimal, language-agnostic contract that standardizes how local...	16	Experimental	1	Shell
101	KonghaYao/ts-sandbox-server A secure, high-performance TypeScript/JavaScript execution sandbox server...	16	Experimental	1	TypeScript
102	RobinhoX/llm-rustyolo 🔒 Run AI agents securely with filesystem, privilege, and network isolation...	16	Experimental	—	Rust
103	ydevil2009/AgentFense 🔒 Enforce least-privilege access for AI agents to safely run untrusted code...	15	Experimental	1	HTML
104	kraaakilo/opencode-vm Isolated Ubuntu VM setup for running OpenCode AI agents safely — Vagrant +...	15	Experimental	1	Shell
105	deepsarda/Nox Nox is a secure, embeddable sandbox runtime for executing untrusted scripts...	15	Experimental	1	Kotlin
106	hanu-tayal/local-agent-sandbox Privacy-first local AI agent runtime: sandboxed execution, sensitivity...	14	Experimental	—	Python
107	aliathiullah/the-agent-sandbox-taxonomy Provide a framework to evaluate AI agent sandboxes by scoring defense layers...	14	Experimental	—	Go
108	ParthSareen/zuko Read-only CLI sandbox for AI Agents with Touch ID for unlocking commands	14	Experimental	—	Go
109	tobocop2/beebox Secure Docker sandbox for running AI coding agents in isolated containers —...	14	Experimental	3	Shell
110	sourcery-zone/agent-vm 🛡️ Security by Compartmentalization for AI Coding Agents.	14	Experimental	3	Nix
111	iamladi/sandcaster Run Pi agents in secure cloud sandboxes — via API, CLI, or Slack. One call....	14	Experimental	—	TypeScript
112	yevhen/klitka Local sandbox runtime for running LLM workloads inside a microVM with...	11	Experimental	—	Go
113	openagentworld/openagentworld-sandbox A framework-agnostic sandbox for AI agent code execution — works with...	10	Experimental	1	Python

Comparisons in this category

E2B and infra (92 vs 76) OpenSandbox and boxlite (87 vs 64) OpenSandbox and nono (87 vs 63) OpenSandbox and infra (87 vs 76) OpenSandbox and agbcloud-sdk (87 vs 54) OpenSandbox and sandboxed.sh (87 vs 45) OpenSandbox and Bandsox (87 vs 53) OpenSandbox and runtm (87 vs 45) OpenSandbox and sandlock (87 vs 57) agent-safehouse and sandboxed.sh (59 vs 45)