The agent harness belongs outside the sandbox
Key takeaways
- Blog The Agent Harness Belongs Outside the Sandbox Andrea Luzzardi·Apr 10, 2026·7 min read.
- They have different security properties, different failure modes, and different implications for what the agent can do.
- The loop lives in the same container as the code it's working on.
Blog The Agent Harness Belongs Outside the Sandbox Andrea Luzzardi·Apr 10, 2026·7 min read. An agent harness is the loop that drives an LLM. It sends a prompt, gets a response, executes the tool calls the model requested, feeds the results back, and repeats until the model says it's done. Every production agent has one. The question is where it runs.
There are two answers. They have different security properties, different failure modes, and different implications for what the agent can do. The tradeoffs also look different depending on whether you're building a single-user agent (one engineer on a laptop) or a multi-user one (dozens of engineers in the same organization sharing the same agent). We're in the multi-user camp, which surfaces problems single-user builders don't hit.
The loop lives in the same container as the code it's working on. LLM calls go out from inside the container. Tool calls (bash, read, write) execute locally. Skills, memories, and anything else the harness tracks are files on the container's filesystem.