Google Research's Harness² lets agents build their own harness per task
thisiskp_ · x · 2026-10-26
The Prompt Engineering YouTube channel posted a 60-second explainer of Harness², a Google Research framework that lets an AI agent build and evolve its own harness — its instructions, tools, and control loop — for every task, with no retraining. All resources are public: paper arXiv:2610.14224, GitHub repo google-research/harness2, and project page harness2.github.io. It's an interesting research direction handing workflow/tool configuration itself over to agent self-optimization.
More from coding & agent
- Ruff creator Charlie Marsh: frontier models find infinite bugs in any software — charliermarsh · 2026-10-26
- Dev testing one skill across Opus 5.5 and Sol 6.1 finds instruction following 'all over the map' — lucasmeijer · 2026-10-26
- Hugging Face uses small open models to reverse engineer a game binary byte-for-byte — yb2698 · 2026-10-26
- ContractSim: first framework evaluating LLM negotiation over contingent natural-language contracts — xuanalogue · 2026-10-26
- OpenAI computer-use agents near superhuman speed, say the team building them — a16z · 2026-10-26
- Qdrant ships one-CLI Chroma migration tool that copies embeddings without re-embedding — qdrant_engine · 2026-10-26