SOURCE-LINKED INTELLIGENCE

Harness-Zero: Harness Distillation via Agent-as-Harness

arXiv · AI, language, vision and robotics · article · Sep 21, 2026 · UTC

Agent harnesses, the external systems that mediate model-environment interaction, can substantially improve agent performance, but their gains remain tied to the harness at deployment. Because the best harness varies across domains, instances, and models, a general-purpose agent must either settle for a suboptimal shared harness or route among an ever-growing set of specialized ones. We therefore study agent harness distillation: using a domain- or instance-optimized harness as training-time guidance and transferring the behaviors it induces into model weights, so that its gains survive under

Read original source ↗ Open in workspace

recordType

paper

region

Global