cs.SEAug 17, 2026

Graphectory Viewer: A Tool for Process-Centric Analysis of Agentic Software Trajectories

Authors: Charlie Jyu, Shuyang Liu, Reyhaneh Jabbarvand

Organizations: University of Illinois at Urbana-Champaign Champaign, USA

Read PDF
Ask AI
Cite

Copy citation

DOINot available for this paper

Abstract

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.

Figures & tables

Figure 1. The Graphectory for a SWE-agent trajectory repairing astropy-13033 . Phase-colored nodes summarize actions, solid edges show execution flow, and dashed blue edges indicate intra-step command links. A phase-colored Graphectory graph for an agent trajectory repairing astropy issue 13033, with nodes for commands and thoughts connected by solid execution edges and dashed intra-step links.
Figure 2. The Phase Sankey Diagram interface with status filtering, phase chips, and sliders for transition depth and minimum displayed flow. Screenshot of the Graphectory Viewer Phase Sankey Diagram interface, showing phase transitions, filters, and a searchable trajectory list.
SourceNNlinesnodesdedup.path
All collections397335,66434.5314.71%16.70
OH Claude-45002,00459.909.60%30.49
OH DeepSeek-V35005,21925.9811.87%13.75
OH DeepSeek-R14743,68119.4611.07%12.57
OH Devstral50012,44672.5310.37%35.66
SA DeepSeek-V349915,89514.9920.71%7.50
Table 1. Trajectory and graph average statistics across released SWE-agent and OpenHands collections.

Explore similar work

CardsList

  1. AgentTrails: Towards Trust and Reuse for Agentic Tasks

    Jul 21, 2026Eden Wu, Sonia Castelo, Yurong Liu +2Data ProvenanceAgent Trajectory Analysis

    New York University

  2. TraceView: Interactive Visualization of Agentic Program Repair Trajectories

    Jun 20, 2026Amirali Sajadi, Tu Nguyen, Kimmie Huynh +2Agent Failure AnalysisSoftware Engineering Agents

    Drexel University, Philadelphia, PA, USA · Belmont University, Nashville, TN, USA

  3. Agent trajectories as programs: fingerprinting and programming coding-agent behavior

    Jun 15, 2026Hamidah OderinwaleAI Coding AgentsAI Agent Evaluation

    Work done with support from Taste Labs