openkruise/agents

agent-sandbox enables easy management of isolated, stateful, singleton workloads, ideal for use cases like AI agent runtimes and reinforcement learning (RL).

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

Runtime-agnostic Kubernetes governance for AI agents: gVisor isolation, egress policy, cost controls, and offline audit verification primitives.

A Kubernetes-native control plane for AI agent instance management, with governed AI access, runtime orchestration, and reusable resources across multiple agent runtimes.

Agent Substrate: the core system

Cloud Native Agentic AI | Discord: https://bit.ly/kagentdiscord

Agent Reference Stack for Kubernetes (kars) - an open source stack from Microsoft for running AI agents safely on Kubernetes. Multi-runtime, Foundry-aware, hardened per-agent sandboxes, governed egress, end-to-end encrypted inter-agent mesh.

Agent-orca orchestrates and manages AI agent workloads in Kubernetes, providing distributed execution, model routing, knowledge integration, and tool access through MCP and stdio tools.

Kubernetes for AI agents: CoralOS is a platform for everything between your agents and production: registry, runtimes, security, and orchestration.