AgentGUI：一个用于观察和引导长时程 AI 智能体的本地 GUI

View PDF HTML (experimental)

Abstract:AI agents are increasingly adept at tackling complex, long-running tasks. With the rapid surge of autonomous capabilities, human oversight is systematically lagging behind due to limited human-centered interfacing. Aiming to address this, we introduce AgentGUI, a user-friendly, locally hosted GUI for seamlessly observing and steering AI agents amid multiple concurrent, long-running sessions. AgentGUI features 1) rich agent trajectory visualizations, 2) effective manual and automated steering, and 3) integration with and coordination between open-source and frontier agent frameworks. A controlled user study demonstrates statistically significant reduction in the time it takes to identify key elements from agent traces (38% faster, p = 0.023). In a preliminary experiment, AgentGUI's automated drift prevention feature raises the task completion rate of small local agents by as high as 34pp across a 0.8B--9B model ladder (N=50 runs per model). AgentGUI is publicly available through its project website (this https URL) and open-source repository (this https URL), along with a demo video (this https URL).

Comments:	Accepted to EMNLP 2026 System Demonstrations
Subjects:	Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Human-Computer Interaction (cs.HC)
Cite as:	arXiv:2607.26300 [cs.CL]
	(or arXiv:2607.26300v3 [cs.CL] for this version)
	https://doi.org/10.48550/arXiv.2607.26300 arXiv-issued DOI via DataCite

Submission history

From: Xuan Zhao [view email]
[v1] Tue, 28 Jul 2026 21:47:22 UTC (2,842 KB)
[v2] Tue, 4 Aug 2026 13:59:52 UTC (2,842 KB)
[v3] Fri, 23 Oct 2026 16:32:34 UTC (2,843 KB)