Skip to main navigation Skip to search Skip to main content

Bugs in Modern LLM Agent Frameworks: An Empirical Study

  • Xinxue Zhu
  • , Jiacong Wu
  • , Xiaoyu Zhang
  • , Tianlin Li
  • , Yanzhou Mu
  • , Juan Zhai
  • , Chao Shen
  • , Chunrong Fang
  • , Yang Liu
  • Nantong University
  • Nanjing University
  • Nanyang Technological University
  • Beihang University
  • University of Massachusetts

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

LLM agents have been widely adopted in real-world applications, relying on agent frameworks for lifecycle execution and multiagent coordination. As these systems scale, understanding bugs in the underlying agent frameworks becomes critical. However, existing work mainly focuses on agent-level failures, overlooking framework-level bugs. To address this gap, we conduct an empirical study of 998 issue reports from CrewAI and LangGraph, and construct a taxonomy of 15 root causes and seven observable symptoms across five agent lifecycle stages: 'Agent Initialization', 'Perception', 'Self-Action', 'Mutual Interaction' and 'Evolution'. Our findings show that agent framework bugs mainly arise from 'API Misuse', 'API Incompatibility', and 'Documentation Desync', largely concentrated in the 'Self-Action' stage. Symptoms typically appear as 'Functional Error', 'Crash', and 'Build Failure', reflecting disruptions to task progression and control flow.

Original languageEnglish
Title of host publicationFSE Companion 2026 - Proceedings of the 34th ACM International Conference on the Foundations of Software Engineering
EditorsShin Hwei Tan, Foutse Khomh
PublisherAssociation for Computing Machinery, Inc
Pages1588-1592
Number of pages5
ISBN (Electronic)9798400726361
DOIs
StatePublished - 17 Jul 2026
EventACM International Conference on the Foundations of Software Engineering, FSE 2026 - Montreal, Canada
Duration: 5 Jul 20269 Jul 2026

Publication series

NameFSE Companion 2026 - Proceedings of the 34th ACM International Conference on the Foundations of Software Engineering

Conference

ConferenceACM International Conference on the Foundations of Software Engineering, FSE 2026
Country/TerritoryCanada
CityMontreal
Period5/07/269/07/26

Keywords

  • bug taxonomy
  • empirical study
  • LLM agent framework

Fingerprint

Dive into the research topics of 'Bugs in Modern LLM Agent Frameworks: An Empirical Study'. Together they form a unique fingerprint.

Cite this