# SelfHeal: Empirical Fix Pattern Analysis and Bug Repair in LLM Agents

arXiv: https://arxiv.org/abs/2604.17699 · HTML: https://mlsanity.com/paper/2604.17699/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Placed in Software engineering coding agents, one of 16 directions in LLM agents and tool use.
- Builds on 15 of its 66 references that are in the map.
- First posted on arXiv 20 Apr 2026.

## Placed in

- LLM agents and tool use / [Software engineering coding agents](https://mlsanity.com/agents/software-engineering-coding-agents/)

## Builds on (most cited first, 15 of 15)

- [LLaMA: Open and Efficient Foundation Language Models](https://mlsanity.com/paper/2302.13971/) (2023)
- [Qwen Technical Report](https://mlsanity.com/paper/2309.16609/) (2023)
- [AgentTuning: Enabling Generalized Agent Abilities for LLMs](https://mlsanity.com/paper/2310.12823/) (2023)
- [Large Language Model-Based Agents for Software Engineering: A Survey](https://mlsanity.com/paper/2409.02977/) (2024)
- [TRAIL: Trace Reasoning and Agentic Issue Localization](https://mlsanity.com/paper/2505.08638/) (2025)
- [Interactive Debugging and Steering of Multi-Agent AI Systems](https://mlsanity.com/paper/2503.02068/) (2025)
- [A Comprehensive Study on Deep Learning Bug Characteristics](https://mlsanity.com/paper/1906.01388/) (2019)
- [DeepLocalize: Fault Localization for Deep Neural Networks](https://mlsanity.com/paper/2103.03376/) (2021)
- [From Language to Action: A Review of Large Language Models as Autonomous Agents and Tool Users](https://mlsanity.com/paper/2508.17281/) (2025)
- [The Future of AI-Driven Software Engineering](https://mlsanity.com/paper/2406.07737/) (2024)
- [MORepair: Teaching LLMs to Repair Code via Multi-Objective Fine-tuning](https://mlsanity.com/paper/2404.12636/) (2024)
- [Proof Repair across Type Equivalences](https://mlsanity.com/paper/2010.00774/) (2020)
- [Anomaly Detection in Large-Scale Cloud Systems: An Industry Case and Dataset](https://mlsanity.com/paper/2411.09047/) (2024)
- [When Agents Fail: A Comprehensive Study of Bugs in LLM Agents with Automated Labeling](https://mlsanity.com/paper/2601.15232/) (2026)
- [The Impact Of Bug Localization Based on Crash Report Mining: A Developers' Perspective](https://mlsanity.com/paper/2403.10753/) (2024)

## Built on by (most cited first, 1 of 1)

- [Diagnosis-Driven Automatic Repair for Agentic Workflow via Symbolic Inference](https://mlsanity.com/paper/2607.02882/) (2026)

## Abstract (arXiv)

Large Language Models (LLMs) have transformed software development and AI applications. While LLMs are designed for text processing, LLM agents extend this capability by enabling autonomous actions, tool use, and multi-step task completion. As this field grows, developers face new challenges in debugging these complex systems. To address this challenge, we present the first empirical study on bug fix patterns in LLM agents. We study buggy posts and code snippets from three platforms: Stack Overflow, GitHub, and HuggingFace Forums. We examine their fix patterns, the components where fixes are applied, and the programming languages and frameworks involved. Furthermore, we introduce AgentDefect, the first benchmark dataset for bugs in LLM agents. The dataset contains 37 runtime buggy instances along with fixed code and test files. Finally, we present SelfHeal, a multi-agent system designed to fix bugs in LLM agents. The system leverages two independent ReAct agents: the fix agent and the critic agent. These agents use tools that provide both internal knowledge (fix rules) and external knowledge (web search) to propose and validate fixes. Our evaluation shows that SelfHeal with Gemini 3 Pro as the backbone LLM outperforms both baseline and state-of-the-art approaches by a significant margin.