TraceSIR: A Multi-Agent Framework for Structured Analysis and Reporting of Agentic Execution Traces

TL;DR

TraceSIR is a multi-agent framework that structures, analyzes, and reports on complex agentic execution traces to facilitate failure diagnosis and root cause analysis, outperforming existing methods.

Contribution

It introduces TraceFormat for trace compression and a multi-agent system for detailed analysis and report generation, advancing automated diagnostics for agentic systems.

Findings

Produces coherent, informative, and actionable reports

Outperforms existing approaches across multiple evaluation metrics

Effectively diagnoses issues in real-world agentic scenarios

Abstract

Agentic systems augment large language models with external tools and iterative decision making, enabling complex tasks such as deep research, function calling, and coding. However, their long and intricate execution traces make failure diagnosis and root cause analysis extremely challenging. Manual inspection does not scale, while directly applying LLMs to raw traces is hindered by input length limits and unreliable reasoning. Focusing solely on final task outcomes further discards critical behavioral information required for accurate issue localization. To address these issues, we propose TraceSIR, a multi-agent framework for structured analysis and reporting of agentic execution traces. TraceSIR coordinates three specialized agents: (1) StructureAgent, which introduces a novel abstraction format, TraceFormat, to compress execution traces while preserving essential behavioral information; (2) InsightAgent, which performs fine-grained diagnosis including issue localization, root cause analysis, and optimization suggestions; (3) ReportAgent, which aggregates insights across task instances and generates comprehensive analysis reports. To evaluate TraceSIR, we construct TraceBench, covering three real-world agentic scenarios, and introduce ReportEval, an evaluation protocol for assessing the quality and usability of analysis reports aligned with industry needs. Experiments show that TraceSIR consistently produces coherent, informative, and actionable reports, significantly outperforming existing approaches across all evaluation dimensions. Our project and video are publicly available at https://github.com/SHU-XUN/TraceSIR.