On the Impact of AGENTS.md Files on the Efficiency of AI Coding Agents

TL;DR

This study investigates how AGENTS.md files influence the efficiency of AI coding agents on GitHub, finding that such files reduce runtime and token usage without compromising task completion.

Contribution

It provides the first empirical analysis of AGENTS.md files' impact on AI coding agents' operational efficiency in real-world repositories.

Findings

AGENTS.md files lower median runtime by 28.64%

AGENTS.md files reduce token consumption by 16.58%

Efficiency gains are achieved without affecting task success rates

Abstract

AI coding agents such as Codex and Claude Code are increasingly used to autonomously contribute to software repositories. However, little is known about how repository-level configuration artifacts affect operational efficiency of the agents. In this paper, we study the impact of AGENTS$.$md files on the runtime and token consumption of AI coding agents operating on GitHub pull requests. We analyze 10 repositories and 124 pull requests, executing agents under two conditions: with and without an AGENTS$.$md file. We measure wall-clock execution time and token usage during agent execution. Our results show that the presence of AGENTS$.$md is associated with a lower median runtime ($\Delta 28.64$%) and reduced output token consumption ($\Delta 16.58$%), while maintaining a comparable task completion behavior. Based on these results, we discuss immediate implications for the configuration and deployment of AI coding agents in practice, and outline a broader research agenda on the role of repository-level instructions in shaping the behavior, efficiency, and integration of AI coding agents in software development workflows.