自我進化編碼代理

摘要

大型語言模型日益被嵌入軟體工程工作流程中，作為能夠檢查程式碼庫、呼叫工具、執行測試、偵錯失敗並產生修補程式的編碼代理。然而，即使軟體開發是一個動態且充滿回饋的過程——程式碼庫會演進、依賴關係會改變、測試會失敗，且修復嘗試會留下可重複使用的經驗——大多數現有代理在部署後仍基本保持靜態。這種緊張關係促成了越來越多的關於自我演化編碼代理的研究，這類代理透過根據先前的編碼互動更新其框架、記憶、技能、工具、模型或協作結構，來改善其未來行為。在本綜述中，我們對這一新興領域進行系統性的綜合整理。我們首先定義自我演化編碼代理，並將其與傳統編碼代理及通用自我演化代理區分開來。接著，我們提出一個以對象為中心的分類法，用以描述這些系統中演化的對象，並輔以兩個正交視角：演化發生的時機，以及哪些軟體特定證據驅動演化。整體文獻顯示，可執行的回饋、程式碼庫級別的上下文與編碼軌跡，使軟體工程在代理自我演化方面成為一個獨特而自然的領域，但同時也在回饋可靠性、基準過擬合、安全性、可維護性、成本與泛化方面引入了新的挑戰。透過圍繞這些維度組織現有工作，本綜述旨在釐清自我演化編碼代理的概念邊界，並為設計更具適應性、可靠性與軟體感知能力的代理系統奠定基礎。我們收集的論文可在 https://github.com/zhouhao1024/Awesome-Self-Evolving-Coding-Agents 查看。

English

Large language models are increasingly embedded in software engineering workflows as coding agents that can inspect repositories, invoke tools, execute tests, debug failures, and generate patches. Yet most existing agents remain largely static after deployment, even though software development is a dynamic, feedback-rich process in which repositories evolve, dependencies change, tests fail, and repair attempts leave reusable experience. This tension has motivated a growing body of work on self-evolving coding agents, where the agent improves its future behavior by updating its framework, memory, skills, tools, models, or collaboration structures from prior coding interactions. In this survey, we provide a systematic synthesis of this emerging area. We first define self-evolving coding agents and distinguish them from conventional coding agents and general self-evolving agents. We then develop an object-centered taxonomy that characterizes what evolves in these systems, and complement it with two orthogonal perspectives: when evolution occurs and what software-specific evidence drives it. Across the literature, we find that executable feedback, repository-level context, and coding trajectories give software engineering a distinctive role as a natural domain for agent self-evolution, but also introduce new challenges in feedback reliability, benchmark overfitting, safety, maintainability, cost, and generalization. By organizing existing work around these dimensions, this survey aims to clarify the conceptual boundaries of self-evolving coding agents and provide a foundation for designing more adaptive, reliable, and software-aware agentic systems. The papers we collect can be found at https://github.com/zhouhao1024/Awesome-Self-Evolving-Coding-Agents.