可视化智能体执行轨迹，揭示行为模式与问题求解过程。
Graphectory Viewer: A Tool for Process-Centric Analysis of Agentic Software Trajectories
- 将多框架智能体轨迹转为分阶段图结构，连接底层执行与高层行为。
- 支持大规模轨迹搜索、过滤及相位转移的桑基图分析。
- 开源工具+预计算图+数据集，助力研究可复现性与行为对比。
我们提出 Graphectory Viewer，一个基于网页的交互式工具，用于对智能体软件轨迹进行以流程为中心的分析。该工具基于此前工作的 Graphectory 表示，将异构原始轨迹转换为包含阶段信息的图结构，连接低层执行细节与高层行为模式。支持来自多个智能体框架的轨迹，提供交互式图构建、节点级思维、动作与观察检查、大规模轨迹集合的搜索与过滤，以及问题求解阶段转换的桑基图摘要。这些功能使研究人员和实践者能够审视单个执行过程，识别重复的行为模式，比较成功与失败运行，并在任务最终结果之外分析大规模轨迹语料库。为支持可复现性和进一步研究，我们开源了 Graphectory Viewer 工具，附带文档、预计算图和大规模轨迹语料库。
原文摘要 · Abstract (English)
We present Graphectory Viewer, a web-based tool for interactive, process-centric analysis of software-agent trajectories. Building on the Graphectory representation introduced in our previous work, Graphectory Viewer transforms heterogeneous raw trajectories into phase-aware graphs that connect low-level execution details with higher-level behavioral structures. The tool supports trajectories from multiple agent frameworks and provides interactive graph construction; node-level inspection of thoughts, actions, and observations; search and filtering over large trajectory collections; and Sankey-style summaries of problem-solving phase transitions. These capabilities enable researchers and practitioners to inspect individual executions, identify recurring behavioral patterns, compare successful and failed runs, and analyze large trajectory corpora beyond final task outcomes. To support reproducibility and further research, we release Graphectory Viewer as an open-source artifact together with documentation, precomputed graphs, and the large-scale trajectory corpus.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。