Search papers, labs, and topics across Lattice.
AgentDebugX is an open-source debugging framework designed to enhance failure observability and recovery in LLM agents by implementing a closed-loop process of Detect, Attribute, Recover, and Rerun. The core method, DeepDebug, utilizes multi-turn root-cause diagnosis through global trajectory understanding and structure-guided investigation, achieving a significant improvement in attribution accuracy on the Who and When benchmark. Additionally, it demonstrates effective task recovery, repairing 13 out of 73 failed tasks in a single rerun, thereby increasing overall accuracy from 55.8% to 63.6%, outperforming existing self-correction methods.
DeepDebug achieves a 32% improvement in task recovery accuracy, showcasing a powerful new approach to debugging LLM agent failures.
LLM agent failures are difficult to debug because the step where an error surfaces is often not the one that caused it. Existing observability tools replay execution traces but provide little support for identifying the root cause or translating diagnosis into recovery. We present AgentDebugX, an open-source debugging framework that organizes debugging as a closed loop of Detect, Attribute, Recover, and Rerun. At its core, DeepDebug performs multi-turn root-cause diagnosis through global trajectory understanding, structure-guided investigation, and cross-examination. On the Who and When benchmark, DeepDebug achieves the best strict attribution accuracy among the evaluated methods on both tested open-weight backbones, reaching 28.8 percent exact agent-and-step accuracy on qwen3.5-9b versus 21.7 percent for the strongest single-pass baseline. On GAIA, DeepDebug repairs 13 of 73 failed tasks in a single rerun, compared with 4 to 6 for three decoupled self-correction baselines, improving overall accuracy from 55.8 percent to 63.6 percent. AgentDebugX exposes this workflow through a Python library, CLI, web console, and installable agentic skill, and provides an opt-in Error Hub for sharing scrubbed failure-diagnosis-repair bundles and reusing them as debugging memory.
Google Research