Skip to content

Awesome list for AI agent harness engineering: tools, patterns, evals, memory, MCP, permissions, observability, and orchestration.

Add Google agent-shell-tools to Security, Sandbox & Permissions section

ai-boostefda77295425
diff --git a/README.md b/README.md
index 1ff4361..2774a1e 100644
--- a/README.md
+++ b/README.md
@@ -475,6 +475,7 @@ Real repositories worth studying — each with a note on *why* it's worth your t
- [Grimlock: Guarding High-Agency Systems with eBPF and Attested Channels](https://arxiv.org/abs/2605.27488) — Roblox's agent guard for high-agency systems: eBPF-forced traffic interception channels every agent-to-agent and agent-to-service call through a guard, post-handshake TLS 1.3 attestation binds identity to the channel, and short-lived scope tokens enforce least-privilege delegation across heterogeneous machines and clouds. The key harness insight is that identity, authorization, provenance, and delegation should be enforced beneath the application layer so existing orchestration code remains untouched.
- [Cloudflare Dynamic Workers](https://blog.cloudflare.com/dynamic-workers/) — V8 isolate-based sandboxing for AI-agent-generated code execution, now in open beta. Isolates start in milliseconds using megabytes of memory — 100x faster and up to 100x more memory-efficient than containers. The sandbox intercepts outbound HTTP requests for credential injection so agent code never touches secrets directly. A fundamentally different architectural option from container-based sandboxes (E2B, Daytona).
- [Kubernetes Agent Sandbox](https://github.com/kubernetes-sigs/agent-sandbox) — K8s-native Sandbox CRD (under SIG Apps) providing declarative, standardized APIs for managing isolated, stateful, singleton workloads for AI agent runtimes. Supports gVisor and Kata Containers for kernel-level isolation; v0.2.1 introduced "Secure by Default" networking architecture enforcing strict isolation with a shared policy model. The right choice when agents must run inside existing Kubernetes infrastructure. ![Stars](https://img.shields.io/github/stars/kubernetes-sigs/agent-sandbox?style=flat-square&label=★&color=yellow)
+- [agent-shell-tools](https://github.com/google/agent-shell-tools) — Google's open-source toolkit for giving coding agents controlled shell access with opinionated defaults: an nsjail-based sandbox, a command-filter rule language, and a gRPC execution proxy so agents can stream commands into isolation without exposing host credentials or filesystems. A concrete reference for separating the agent's shell surface from the host boundary rather than relying on prompt-level trust. ![Stars](https://img.shields.io/github/stars/google/agent-shell-tools?style=flat-square&label=★&color=yellow)
- [Why sandboxing your agent is not enough](https://www.cncf.io/blog/2026/07/07/why-sandboxing-your-agent-is-not-enough/) — CNCF's July 2026 argument that container isolation is necessary but insufficient for production agents. The agent-substrate pattern decouples the agent actor from pod lifecycle so agents execute in short bursts, suspend when idle, and resume on any worker while keeping gVisor/Kata-level isolation. A concrete pattern for high-density, low-idle-cost secure agent fleets in Kubernetes.
- [Alibaba OpenSandbox](https://github.com/alibaba/OpenSandbox) — General-purpose sandbox platform for AI agents (8.7K+ stars, March 2026) with multi-language SDKs (Python, Java, TypeScript, Go, C#), unified APIs across Docker/Kubernetes runtimes, and support for secure container runtimes (gVisor, Kata Containers, Firecracker). Covers coding agents, GUI agents, agent evaluation, and RL training in a single abstraction layer — the most runtime-flexible sandbox option when you need to choose isolation levels per workload. ![Stars](https://img.shields.io/github/stars/alibaba/OpenSandbox?style=flat-square&label=★&color=yellow)
- [CubeSandbox](https://github.com/TencentCloud/CubeSandbox) — Tencent Cloud's production-validated microVM sandbox for AI agents: sub-60ms cold start via snapshot cloning, <5MB per-instance overhead, and true kernel-level isolation with eBPF-enforced network policies. E2B-compatible drop-in replacement that demonstrates how hyperscale cloud infrastructure can be repurposed for high-density agent execution. ![Stars](https://img.shields.io/github/stars/TencentCloud/CubeSandbox?style=flat-square&label=★&color=yellow)