Overeager Coding Agents: Measuring Out-of-Scope Actions on Benign Tasks

Abstract

Coding agents now run autonomously with shell, file, and network privileges. When a user issues a benign request, the agent sometimes does more than asked: it deletes unrelated files, wipes a stale credentials backup, or rewrites configuration the user never mentioned. We call these scope expansions overeager actions, an authorization problem distinct from capability failures, prompt injection, or sandbox escapes. We present OverEager-Gen, a benchmark dedicated to overeager behavior on benign ta

Categories

Framework mappings

Suggested from the entry's categories.

@misc{qu2026overeager,
  title = {{Overeager Coding Agents: Measuring Out-of-Scope Actions on Benign Tasks}},
  author = {Yubin Qu and Ying Zhang and Yanjun Zhang and Gelei Deng and Yuekang Li and Leo Yu Zhang and Yi Liu},
  year = {2026},
  month = may,
  eprint = {2605.18583},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2605.18583}
}