LLM vs Agent vs Harness: X Debate Erupts Over Whether Scaffolding, Not Models, Drives Capability Gains
sean_a_mcclure · x · 2026-09-25
A technical argument between seanamcclure and ryunuck over whether Claude's agentic feats come from the model or the harness. ryunuck argues benchmarks are hugely harness-dependent and older models would score higher with new scaffolding. seanamcclure counters that calling LLMs pattern recognizers sets no upper bound, that different models in the same harness still differ, and that the two are conflating LLM, agent, and AI system.
Related event: Debate: does Claude's power come from model or harness?(3 posts)→
More from coding & agent
- Kaigen, a C-based AI-native game engine, opens closed beta — gdechichi · 2026-09-25
- One-prompt Minecraft: AI-generated voxel game open-sourced, runs in browser and on Windows — gdechichi · 2026-09-25
- He used $100 of Claude credits to land his first VSCode PR — ThePeterMick · 2026-09-25
- What the OpenAI-Hugging Face incident says about agent oversight — rainerhahnekamp · 2026-09-25
- Podcast on AI state and harness engineering, with all visuals generated by Opus 5.5 — tensorqt · 2026-09-25
- I trained an AI on 25 years of my writing as a reader, not a mimic — UnionPacifik · 2026-09-25