Research: AGENTS.md File Can Reduce AI Agent Runtime by 29% and Output Token by 17%

According to 1M AI News, a research team from the Singapore Management University, Heidelberg University, Bamberg University, and King's College London has published a paper on arXiv, quantitatively assessing for the first time the impact of repository-level configuration file AGENTS.md on the efficiency of AI programming Agents. AGENTS.md is an instruction file stored in the root directory of a code repository, used to guide the AI Agent on project architecture, build commands, coding standards, and operational constraints, similar to Anthropic's CLAUDE.md and GitHub Copilot's copilot-instructions.md. It is currently adopted by over 60,000 GitHub repositories.

The research team conducted paired experiments on 124 merged PRs (each with code changes of no more than 100 lines) in 10 repositories using OpenAI Codex (gpt-5.2-codex), running under two conditions: with and without AGENTS.md. The results showed that the median runtime decreased from 98.57 seconds to 70.34 seconds (a reduction of 28.64%) when AGENTS.md was present, the median output tokens decreased from 2,925 to 2,440 (a reduction of 16.58%), with no significant difference in task completion behavior (Wilcoxon signed-rank test, p <0.05).

The researchers noted that AGENTS.md transforms Agent guidance from a "brief prompt" to a "version-controlled, reviewable, collaboratively maintained configuration artifact," and suggested that development teams adopt it as a standard practice in their repositories. Limitations of the study include testing only a single Agent with OpenAI Codex, a sample limited to small-scale PRs, and a lack of comprehensive code correctness evaluation.