Augment Code: A Good AGENTS.md Is a Model Upgrade, a Bad One Is Worse Than No Docs — 125 Points on HN

Augment Code / Hacker News·medium signal

Augment Code published research showing that 100-150 line AGENTS.md files with focused reference docs delivered 10-15% improvements across all coding metrics, but longer files reversed gains. Key insight: LLM-generated context files hurt performance in 5 of 8 tested settings, increasing inference costs 20-23% and adding 2.5-3.9 extra steps per task. Procedural numbered workflows were the strongest pattern — moving agents from unable to first-try correct. 125 points, 42 comments.