一句话：提出以真实开发实践为基础的 LLM 智能体评估方法论，涵盖污染感知、真实场景智能体行为与轨迹级基准指标。
大语言模型正从辅助工具演变为深度嵌入协作开发环境的自主贡献者，但现有评估技术碎片化，且多依赖假想的语法化场景，扭曲了模型真实能力的投影。本文提出一套面向 LLM 驱动的软件工程智能体的综合评估方法论，扎根于真实软件开发实践。其核心包含三个方面：数据污染感知（contamination-awareness），确保评估不被训练数据泄漏所污染；in-the-wild 的智能体行为评估，在真实编码情境中考察智能体的自主行为；以及轨迹感知（trajectory-aware）的基准与指标，用以刻画真实编码上下文、与人类对齐的行为以及模型的失效模式。该工作属于研究愿景与方法论层面的贡献，旨在为智能体评估提供更可靠、更贴近开发者实际需求的框架。
Large language models are rapidly moving towards closing the development cycle, transitioning from simple assistive companions to autonomous contributors deeply embedded into collaborative development environments. Despite their accelerated adoption, existing evaluation techniques are limited due to their fragmented nature and distorted projection of true model capabilities, often obtained from hypothetical syntactic scenarios. This research aims to bridge this gap by providing a comprehensive evaluation methodology for LLM-powered agents that is grounded in real-world software development practice. Our evaluation approach focuses on contamination-awareness, in-the-wild agentic behavior assessment, and trajectory-aware benchmarks and metrics capturing realistic coding contexts, human-aligned behavior, and model failure modes.