Using sandboxes to stop agents from cheating
· Rotem Tamir
Introducing RewardHackBench, an open-source benchmark for measuring whether sandbox policies prevent reward hacking in agent evaluations.
Blog
Blog
Agent safety, isolation, and building with Islo.
When you should use Kubernetes Agent Sandbox
· Adam Gold
A technical look at what Agent Sandbox solves inside Kubernetes, what it leaves to the platform team, and how to decide if it's the right primitive for your setup.
Coding Agents Need a New Primitive, and It's Not a Sandbox
· Adam Gold
Fast boot times are useful for request-shaped compute. Coding agents need something closer to a secure, persistent, workstation-shaped runtime.
What 100 Conversations Taught Us About Autonomous Coding Agents
· Adam Gold
A framework for evaluating what you need to run AI coding agents at scale and where most teams get stuck.
Islo vs Claude Code for Web: Different Tools for Different Jobs
· Adam Gold
Claude Code for Web is great for quick coding help in the browser. Islo lets you run Claude Code alongside Cursor, orchestrate multiple agents, and connect to real services - all unattended.
Islo vs Cursor Cloud Agents: Different Tools for Different Jobs
· Adam Gold
Cursor Cloud Agents is great for quick edits in the browser. Islo lets you run Cursor alongside Claude Code, orchestrate multiple agents, and connect to real services - all unattended.
The missing layer of the agent development stack
· Adam Gold
Stateful fake servers for GitHub, Slack, and Stripe that work with official SDKs - built for AI agent development.
Monty and Islo: Sandbox the Snippet, Isolate the Agent
· Yossi Eliaz
Monty sandboxes runtime-generated Python. Islo gives coding agents a zero-setup environment. Here's what Monty blocks, what can still crash, and how the layers fit.
Best 5 Sandboxes for Claude Code
· Adam Gold
Here are five sandbox options that work with Claude Code (or with AI coding agents in general), and how they compare.
Understanding Sandbox Security
· Adam Gold
Why sandboxing matters for AI agents - isolation, boundaries, and blast radius so you can run agents unattended without risking credentials or production.
You shouldn't run Clawdbot on your computer
· Adam Gold
How I sent my own database credentials to a server I control using Clawdbot - and why prompt injection and full machine access make that possible in about 20 minutes.