ML//agent//agent harness

An LLM can emit text or a structured function call. It cannot open a file, retain durable state, or deploy software by itself. The agent harness is the ordinary software that turns those generations into a process acting on an environment.

An LLM can emit text or a structured function call. It cannot open a file, retain durable state, or deploy software by itself. The agent harness is the ordinary software that turns those generations into a process acting on an environment.

It assembles context, exposes tools, validates calls, enforces permissions, executes implementations, serializes observations, handles failure, and starts the next inference. The model proposes a transition; the harness decides whether and how that transition becomes real.

This is wider than a system prompt and narrower than the whole product. A prompt supplies information. An API supplies operations. The harness owns the loop connecting inference, operations, state, and policy enforcement.

The model is not the agent's operating system. The harness is.