Sandbox Python Packages

cuga

CUGA is an open-source generalist agent harness for the enterprise, supporting complex task execution on web and APIs, OpenAPI/MCP integrations, composable architecture, reasoning modes, and policy-aware features.

pydantic-ai-backend

File Storage & Sandbox Backends for Pydantic AI: console tools for file operations, Docker-isolated sandboxes for safe execution, and permission system with presets for access control. Enables secure multi-user handling and testing in agents via in-memory, local, or containerized storage.

tensorlake

Tensorlake is a serverless runtime for sandboxes and deploying background agentic applications

opensandbox

Secure, Fast, and Extensible Sandbox runtime for AI agents.

evalidate

Safe and fast evaluation of untrusted user-supplied python expressions

wuying-agentbay-sdk

The Cloud Sandbox Built for AI Agents

polyaxon

AI Infra / AI Orchestration / AI Control Plane

microsandbox

🧱 easy, fast and local-first microVM runtime

containarium-telemetry

Open-source agent runtime — SSH-native isolation, eBPF egress policy, Kubernetes + LXC backends, GPU passthrough, MCP-native CLI

opensandbox-code-interpreter

Secure, Fast, and Extensible Sandbox runtime for AI agents.

codeboxapi

👾📦 CodeBoxAPI is the simplest sandboxing infrastructure for your LLM Apps and Services.

boxlite

The micro-VM for AI agents — light enough to embed on your laptop, elastic enough to power an agentic cloud.

agent-airlock

A deny-by-default contract & type-checker layer for AI agent tool calls — Pydantic-based, in-process, zero-core-deps. Validates the actual tool-call payload (ghost-arg stripping, strict types, self-healing retries) beneath MCP gateways & firewalls. Works with LangChain, OpenAI Agents SDK, PydanticAI & CrewAI.

opensandbox-server

Secure, Fast, and Extensible Sandbox runtime for AI agents.

smolvm-core

Open-source AI sandbox infrastructure with unified API for VMMs -- Firecracker, QEMU and libkrun.

repo-proofer

Find out if a repo will steal your keys before you run it. Zero-network sandbox + strace exfil detection. 100% deterministic, no AI. `uvx repo-proofer <url>

codefreedom

Unified interface for all code agents — LLM proxy routing, Docker sandboxing, profile management, and browser tools. Switch models and providers without reconfiguring.

zapcode

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

zu-core

An opinionated, backend-agnostic runtime for agents that work in production — deterministic, auditable, injection-resistant.

vpod

Lightweight, secure linux sandboxes for untrusted processes.

flowent

A workflow orchestration platform for multi-agent collaboration

rhea

ML/AI training/serving and agent sandbox operator and controller for Kubernetes

zu-backends

An opinionated, backend-agnostic runtime for agents that work in production — deterministic, auditable, injection-resistant.

zu-cli

An opinionated, backend-agnostic runtime for agents that work in production — deterministic, auditable, injection-resistant.