Google open-sources Harness², an agent harness that improves itself when it falls short
burny_tech · x · 2026-10-20
Google Research open-sourced Harness² (arXiv:2610.14224), a plug-and-play framework for recursive agent harnessing: when the agent hits problems its harness wasn't equipped for, the harness improves itself — with model weights frozen.
- Task-level recursion: contrastive Propose–Probe–Compose refines the harness for the current task without ground-truth verifiers
- Harnessing-level recursion: extracts lessons from each edit into a persistent improvement harness for future tasks
A single step consistently improves GLM-5 and Gemini 3.5 Flash over their base harnesses: up to +11.5 points on WorkBuddy-Bench, +4.5 on JobBench, +11.8 pts on LAB.
Related event: Google Open-Sources Harness², a Recursively Self-Improving Agent Framework(3 posts)→
More from coding & agent
- Hugging Face ships gr.workflow and calls on the open-source community to build with it — erhannah · 2026-10-20
- TraceVista: An Open-Source Tool for Visualizing and Understanding Agent Trajectories — _AustinCalvert_ · 2026-10-20
- Why GRPO collapses on multi-turn agent training, and how to stabilize advantage estimation — burkov · 2026-10-20
- Clairvoyance turns a C++ strategy game into a pixel-perfect web app for rapid idea testing — draginol · 2026-10-20
- Claude Code finds hidden camera in Airbnb smoke detector via Wi-Fi scan in 2 minutes — 4KTV · 2026-10-20
- SF meetup 'Are We Running Too Many Agents?' hits capacity with exe, Amp, DatologyAI speakers — josh_wills · 2026-10-20