Google Research's Harness² lets agents rewrite their own harness mid-task

Harness2 by Google Research – Recursive Agent Harnessing

Harness² is a plug-and-play framework from Google Research for recursive agent harnessing in an open world, keeping model weights fixed. Task-level recursion runs a contrastive Propose–Probe–Compose step that edits the harness, tests the changes, and composes a refined version from behavioral contrasts without ground-truth feedback; harnessing-level recursion carries adopted harnesses and lessons forward. A single step lifts GLM-5 and Gemini 3.5 Flash by up to 11.5 points on WorkBuddy-Bench, 4.5 on JobBench, and 11.8 percentage points on LAB's partial pass rate.