Harness engineering
Harness engineering is the work of building the system around a model.
In software engineering, a test harness is code written to exercise other code before the code that will eventually call it exists. A June 2026 survey of agent research credits the software engineer Mitchell Hashimoto and OpenAI with making harness engineering a settled term.
Brand rules, approved claims, channel specs, and the test set need no engineers. Connectors, code checks, permissions, and records need engineering time or a vendor.
Whenever an issue happens more than once, the controls that steer and check the agent should be improved to make it less likely or prevent it. Loop 6 does this by changing how the harness works: its skills, checks, and limits.
Example
Anthropic reports a coding experiment where the full harness cost over 20 times as much as one agent working alone, and the lone agent's app had a central feature that didn't work. Its newer model, Claude Opus 4.6, then made the checking agent in that experiment unnecessary for many tasks, though it still helped on the hardest ones.