TRAJDEBUG: Tracing Error Lifecycle to Identify Critical Failures in Long-Horizon Agent Trajectories

Source

• Source name: arXiv
• Source kind: paper
• Authors / organization: Yunjia Qi, Zehua Yin, Xintong Shi, Hao Peng, Songyuanyi Lu, Yixian Liu, Richeng Xuan, Yuhong Liu, Zhichao Hu, Xiaozhi Wang, Lei Hou, Bin Xu, Juanzi Li
• Venue: arXiv: cs.AI
• Published: 2026-08-06
• Collected: 2026-08-07
• URL: http://arxiv.org/abs/2608.06346v1
• PDF: https://arxiv.org/pdf/2608.06346v1
• DOI: UNKNOWN

Plain-Language Brief

The paper proposes TrajDebug, a framework designed to trace errors in long-term AI agent trajectories. The framework addresses challenges in identifying individual errors in long trajectories and distinguishing critical errors from multiple local errors. It uses multi-granularity history compression and evidence-based error identification for error discovery and traces each error's resolution status and terminal impact for critical attribution. The paper also presents TrajErrBench, a benchmark of failed trajectories, and demonstrates that TrajDebug outperforms existing baselines.

What the Document Says

• The paper introduces TrajDebug, an error-lifecycle tracing framework for long-term AI agent trajectories.
• TrajDebug addresses the challenges of identifying individual errors in long trajectories and distinguishing critical errors from multiple local errors.
• It uses multi-granularity history compression and evidence-based error identification for error discovery.
• For critical attribution, it traces each error's resolution status and terminal impact.
• The paper also presents TrajErrBench, a benchmark of failed trajectories.
• Experiments show that TrajDebug outperforms existing baselines.
• The authors will release the codes and data for further research.

Key Concepts Mentioned

• AI agents
• error lifecycle
• long-term trajectories
• critical error detection
• multi-granularity history compression
• evidence-based error identification
• TrajDebug
• TrajErrBench

Claims Made by the Document

• TrajDebug addresses challenges in long-trajectory error discovery.
• Experiments show that TrajDebug outperforms existing baselines.

Evidence Used

• The paper presents TrajDebug as a solution to the challenges faced in long-term AI agent trajectories.
• Results from experiments across diverse agent benchmarks are provided to demonstrate TrajDebug's superior performance.

Methods / Approach

UNKNOWN

Limitations Stated by the Authors

UNKNOWN

Limitations Noted by the Agent

Agent interpretation — not from the source.

UNKNOWN

Open Questions

UNKNOWN

Notes for Future Q&A

May include interpretation. Labeled where it does.

UNKNOWN

Raw Abstract / Excerpt

LLM-based agentic systems have shown remarkable capabilities in complex domains, while suffering from cascading errors and difficulty in debugging. Critical error detection aims to locate the earliest error step in a failed trajectory that is responsible for the final failure. However, progress faces two main challenges. First, long trajectories make it difficult to identify individual errors, since the evidence for judging a step may be scattered across distant instructions, observations, and prior context. Second, failed trajectories often contain multiple local errors with different downstream effects, only some of which remain responsible for the final failure. In this work, we propose TrajDebug, an error-lifecycle tracing framework that addresses long-trajectory error discovery with multi-granularity history compression and evidence-based error identification, and supports critical attribution by tracing each error's resolution status and terminal impact. We further construct TrajErrBench, a benchmark of 486 manually annotated failed trajectories from Tau2Bench and SWE-Bench Pro, covering realistic tool-use and coding scenarios. Experiments across diverse agent benchmarks show that TrajDebug achieves the best overall performance over existing baselines, and application studies further demonstrate that its diagnoses provide actionable feedback for improving downstream agent success. We will release the codes and data to facilitate further research.