Tool Count Doesn't Predict Agent Gains: Correlation Is Just −0.15
yb2698 · x · 2026-10-24
In a follow-up to the harness2 reproduction thread, the author shows tool count doesn't predict gains: correlation between tools added and runs gained is −0.15. floor-plan got 3 tools and a 452-word prompt yet lost 5 runs; protein-sequence gained 7 runs from just 1 tool. The right tool matters, not the number.
Related event: Reproducing Google's harness2: Qwen3-9B pass rate jumps from 31% to 45%(6 posts)→
More from coding & agent
- Claude approved its own broken 3D screenshots — how a 480Hz verification gate fixed agent self-grading — youthanasia669 · 2026-10-24
- PM's step-by-step guide to building product strategy with Claude Code in 2 hours — aakashgupta · 2026-10-24
- Six commonly confused LLM API terms: context window, temperature, reasoning tokens — ClickOk5811 · 2026-10-24
- Mubit trains proprietary models from your agent traces so you stop being an API tenant — SucceededMind · 2026-10-24
- Microsoft Open-Sources Frontier Tour Resource Hub with Code, Demos, and Labs — WirelessLife · 2026-10-24
- Open-source kapitsa engine boosts Mac local agent throughput 23% with tree speculative decoding — DotaMate · 2026-10-24