Google Research open-sources Harness², a recursive agent harnessing framework (up to +11.5 pts)
Saboo_Shubham_ · x · 2026-10-18
Google Research released Harness² (Apache-2.0, 1.8k stars), a plug-and-play framework for recursive agent harnessing with frozen model weights. Task-level recursion refines the execution harness via a contrastive Propose–Probe–Compose loop, while harnessing-level recursion maintains a persistent improvement harness with evidence-backed lessons. A single step lifts GLM-5 and Gemini 3.5 Flash over their base harnesses by up to +11.5 pts on WorkBuddy-Bench, +4.5 on JobBench, and +11.8 pts partial pass on LAB. Supports parallel/sequential recursion, OpenCode and Codex executors.
More from coding & agent
- Hermes Agent + local models manage all machines; hermes-skins repo hits 607 stars — Teknium · 2026-10-19
- Harness is dead? AI Engineer NYC debates stripping scaffolds vs. multi-day agents — TheTuringPost · 2026-10-19
- Open-source Claude Code plugin routes grunt work to cheap models, saves Opus for judgment — investigatormaker · 2026-10-19
- Developer builds a standalone Mac app from scratch in 1.5 hours with Cursor — gaganghotra_ · 2026-10-19
- Apify founder Jan Curn: every agent payment protocol exists, real volume is near zero — AI Engineer · 2026-10-19
- How I imported 10,000+ media notes into Obsidian with AI, safely — dSebastien · 2026-10-19