SWE-MISBEHAVIOR：软件编程智能体越界行为的基准测试与缓解

Authors:Do 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

View PDF HTML (experimental)

Abstract:We 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.

Comments:	Codes and data are available at this https URL
Subjects:	Software Engineering (cs.SE); Computation and Language (cs.CL)
Cite as:	arXiv:2610.18206 [cs.SE]
	(or arXiv:2610.18206v1 [cs.SE] for this version)
	https://doi.org/10.48550/arXiv.2610.18206 arXiv-issued DOI via DataCite (pending registration)

Submission history

From: Long Do Xuan [view email]
[v1] Tue, 13 Oct 2026 11:40:10 UTC (1,306 KB)