# Harness²: Google's Recursive Agent Harnessing Improves Frozen Models by Up to 11.8 Points > Source: > Published: 2026-10-19 00:00:00+00:00 # Harness²: Google's Recursive Agent Harnessing Improves Frozen Models by Up to 11.8 Points There are two ways to make an AI agent smarter: change the model, or change everything around the model. Retraining weights is expensive, slow, and often out of reach. But the *harness* — the prompts, tools, retries, and scaffolding that wrap the model — is software you control. [Harness²](https://github.com/google-research/harness2), from Northwestern University and Google Cloud AI Research, makes a compelling case that the harness is where a lot of the untapped performance actually lives. And it gets there without touching a single weight. ## The Problem: Self-Improvement Breaks in the Open World Recursive self-improvement (RSI) — an agent iteratively rewriting its own scaffolding — is a seductive idea. We’ve written about it from several angles: [Anthropic’s claim that Claude writes 80% of its own code](https://themenonlab.blog/anthropic-recursive-self-improvement-ai-builds-itself-june-2026/), [SIA updating its own weights](https://themenonlab.blog/sia-self-improving-ai-agent/), and [Meta-Harness, the agent that rewrites its own scaffolding](https://themenonlab.blog/meta-harness-self-evolving-agent-scaffolding/). The dream is an agent that gets better the more it works. The paper identifies two reasons RSI tends to fall apart in the real (open) world: 1. **No ground-truth verifier.** Real tasks are unseen at design time and novel at test time. Without an oracle telling you whether an edit helped, online self-improvement can happily optimize itself into nonsense. 2. **It doesn’t generalize.** Most RSI is painfully task-specific — tune the harness for task A, learn nothing transferable for task B. ## The Fix: Two Levels of Recursion Harness² (“Harness-squared”) answers both with two connected recursion loops. **Task-level recursion** refines the harness for the *current* task using a contrastive **Propose-Probe-Compose** step: - **Propose** candidate edits to the harness. - **Probe** them*contrastively* — compare candidates against each other to infer which is better, sidestepping the missing ground-truth verifier entirely. - **Compose** the winning changes into an improved harness. This loop can be scaled with a **recursion budget**, run either in parallel or sequentially, and performance keeps climbing as you spend more of it. **Harnessing-level recursion** is the part that makes it generalize: it extracts *experience* from those task-level edits and folds it into a **continual improvement harness**, so lessons learned on one task make the agent better at future, never-seen tasks. That’s the “²” — recursion on the task, and recursion on the harnessing process itself. Crucially, it’s **plug-and-play**: Harness² sits on top of an existing agent harness rather than replacing it. ## The Numbers Across three professional benchmarks and four model–harness configurations, **a single step** of Harness² consistently beat the base harnesses — with **frozen weights**: - **+11.5 points** on WorkBuddy-Bench - **+4.5 points** on JobBench - **+11.8 percentage points** on LAB And these are the *single-step* gains — the paper reports further improvement as the task-level recursion budget grows. Both GLM-5 and Gemini 3.5 Flash benefited, which suggests the method isn’t overfit to one model family. ## Why This Matters The quiet thesis here lines up with something we keep coming back to: for a lot of real work, [you probably don’t need a bigger model](https://themenonlab.blog/small-models-big-performance-nature-paper/) — you need a better harness. [Superpowers made coding agents act like senior developers](https://themenonlab.blog/superpowers-turn-coding-agents-senior-developers/) purely through scaffolding. [Everything Claude Code cut costs 60%](https://themenonlab.blog/everything-claude-code-harness-system/) the same way. Harness² formalizes that intuition into a recursive, verifier-free procedure and shows double-digit gains on professional benchmarks — all at inference time, no fine-tuning, no GPU bill. The verifier-free contrastive probing is the genuinely clever bit. “How do you know an edit helped when there’s no answer key?” has been the Achilles’ heel of open-world self-improvement. Answering it by comparison instead of absolute scoring is a neat trick that others will likely borrow. It’s open-source, it’s plug-and-play, and it targets the layer most teams can actually change. If you’re running agents in production, this is one worth reading closely. **Paper:** [arXiv:2610.14224](https://arxiv.org/abs/2610.14224) · **Code:** [github.com/google-research/harness2](https://github.com/google-research/harness2) · **Project page:** [harness2.github.io](https://harness2.github.io)