titleSWE-MISBEHAVIOR: Benchmarking and Mitigating Boundary Violations in Software Coding Agents
authorsDo Xuan Trong, Duc Anh Nguyen, Do Xuan Long, Ngoc-Hai Nguyen, Tongyao Zhu, Duy Dinh, Cayden Chik, Cuong Dang, Jiaying Wu, Xuan-Phi Nguyen, Vinh Van Nguyen, Kenji Kawaguchi, Shafiq Joty, Nancy F. Chen, Min-Yen Kan
date2026-10-13
venuearXiv (preprint, not peer reviewed)
idarXiv:2610.18206
affiliationNational University of Singapore
upvotes14 (Hugging Face Daily Papers, 2026-11-01)
linksabs pdf hf code
abstractWe present SWE-MISBEHAVIOR, the first taxonomy and evaluation framework for diagnosing boundary-violating misbehaviors in coding agents. It contains 7.4K human-in-the-loop synthesized samples spanning three layers of agent action space: (i) Code Generation, (ii) File Management, and (iii) System Interaction. Across these categories, SWE-MISBEHAVIOR covers 11 distinct misbehaviors with a fully automated evaluation protocol, including coding-action violations, Create/Read/Update/Delete (CRUD) file misbehaviors, system integrity and abuse violations, and task-goal misalignment. Experiments on state-of-the-art open-source coding agents (e.g., OpenHands, smolagents) reveal substantial misaligned behaviors, including dependency misinterpretation (up to 98.5% of cases), unintended modification of user inputs (50.5%), and silent compromise of environment integrity (53%). Motivated by these findings, we further propose CODE-KIT, a lightweight adaptive framework that combines misbehavior-aware guidelines with novel tools to help agents avoid such failures. CODE-KIT reduces misbehaviors by 52% for smolagents and 34% for OpenHands, while simultaneously improving task completion performance.

cite

@misc{trong2026swe,
  title={SWE-MISBEHAVIOR: Benchmarking and Mitigating Boundary Violations in Software Coding Agents},
  author={Do Xuan Trong and Duc Anh Nguyen and Do Xuan Long and Ngoc-Hai Nguyen and Tongyao Zhu and Duy Dinh and Cayden Chik and Cuong Dang and Jiaying Wu and Xuan-Phi Nguyen and Vinh Van Nguyen and Kenji Kawaguchi and Shafiq Joty and Nancy F. Chen and Min-Yen Kan},
  year={2026},
  eprint={2610.18206},
  archivePrefix={arXiv},
  url={https://arxiv.org/abs/2610.18206}
}