Empowering Autonomous Debugging Agents with Efficient Dynamic Analysis

2026-04-27

AI Summary

This work addresses the limitations of existing program repair agents, which rely on coarse-grained execution feedback, and the high cost and incompatibility of traditional line-level debuggers with large language models (LLMs). To overcome these challenges, the authors propose an Agent-oriented Debugging Interface (ADI) that introduces a novel frame lifecycle trace data structure and function-level dynamic analysis. This approach elevates the interaction granularity from line-level to function-level and integrates high-level navigation commands with LLMs to enable efficient end-to-end debugging. Evaluated on SWE-bench Verified, ADI achieves a 63.8% resolution rate at a cost of $1.28 per task, outperforming Claude-Tools. Furthermore, as a plug-and-play module, ADI consistently enhances state-of-the-art agents by 6.2%–18.5% in performance.

Technology Category

• Multiagent Systems: Adversarial Agents
• Planning, Routing, and Scheduling: Replanning and Plan Repair
• Knowledge Representation and Reasoning: Diagnosis and Abductive Reasoning

Application Category

• Search and Retrieval-Augmented AI: Agentic search
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Economics, Online Markets and Human Computation: Cost models of using LLMs in production systems

Abstract

Autonomous agents for automated program repair represent a promising frontier in software engineering, yet their effectiveness is often hindered by reliance on post-mortem, coarse-grained execution feedback. While integrating traditional interactive debuggers seems a natural solution, their low-level, line-by-line interaction paradigm turns out to be cost-inefficient for LLM-based agents, leading to exhausted budgets and unproductive loops. To mitigate this, we introduce Agent-centric Debugging Interface (ADI), a novel agent-centric debugging interface designed for cost-efficient, end-to-end autonomous interaction. Specifically, Agent-centric Debugging Interface realizes a function-level interaction paradigm, powered by our Frame Lifetime Trace, a comprehensive data structure encapsulating a function's stateful execution trace, and a set of high-level navigational commands. Our extensive evaluation on the SWE-bench benchmark demonstrates the effectiveness and efficiency of ADI. By simply equipping a basic agent with ADI, it successfully resolves 63.8\% of the tasks on the SWE-bench Verified set, even slightly outperforming the highly optimized and high-investment Claude-Tools agent, at an average cost of USD 1.28 per task with Claude-Sonnet-3.7. Furthermore, we demonstrate ADI's generality by integrating it as a plug-and-play component into existing SOTA agents, delivering consistent gains ranging from 6.2\% to 18.5\% on the resolved tasks. These results indicate that Agent-centric Debugging Interface can provide a general and efficient enhancement for existing autonomous agents.

Problem

• autonomous debugging
• program repair
• dynamic analysis
• LLM-based agents
• debugging interface

Innovation

• Agent-centric Debugging Interface
• Function-level Interaction
• Frame Lifetime Trace
• Autonomous Program Repair
• Dynamic Analysis