让AI生成真正可用的全栈网页应用，解决代码自动生成中的数据与存储难题。
FullStack-Agent: Enhancing Agentic Full-Stack Web Coding via Development-Oriented Testing and Repository Back-Translation
- 构建多智能体框架，实现前端、后端与数据库的全流程代码生成与调试。
- 通过反向翻译开源仓库提升大模型能力，使30B模型性能提升最高9.7%。
- 推出综合性测试基准，全面评估生成网站的前后端及数据库功能。
让非专业用户借助基于大模型的代码代理开发复杂交互式网站已成为热门任务。然而，现有代码代理多仅生成前端页面，掩盖了真实全栈数据处理与存储的缺失，仅呈现华丽视觉效果。构建生产级全栈网页应用远比单纯生成前端更难，需精确控制数据流、理解持续更新的包依赖，并准确定位代码库中的隐蔽错误。为此，我们提出FullStack-Agent，一个统一的全栈智能体系统，包含三部分：(1) FullStack-Dev，具备强规划、代码编辑、代码库导航和错误定位能力的多智能体框架；(2) FullStack-Learn，一种创新的数据扩增与自我提升方法，通过反向翻译爬取与合成的网站仓库来优化FullStack-Dev的基座大模型；(3) FullStack-Bench，一套系统性测试前后端与数据库功能的综合基准。实验表明，FullStack-Dev在前端、后端和数据库测试用例上分别优于前代最佳方法8.7%、38.2%和15.9%。此外，FullStack-Learn使30B模型在三类测试中性能提升9.7%、9.5%和2.8%，验证了该方法的有效性。代码已开源：https://github.com/mnluzimu/FullStack-Agent。
原文摘要 · Abstract (English)
Assisting non-expert users to develop complex interactive websites has become a popular task for LLM-powered code agents. However, existing code agents tend to only generate frontend web pages, masking the lack of real full-stack data processing and storage with fancy visual effects. Notably, constructing production-level full-stack web applications is far more challenging than only generating frontend web pages, demanding careful control of data flow, comprehensive understanding of constantly updating packages and dependencies, and accurate localization of obscure bugs in the codebase. To address these difficulties, we introduce FullStack-Agent, a unified agent system for full-stack agentic coding that consists of three parts: (1) FullStack-Dev, a multi-agent framework with strong planning, code editing, codebase navigation, and bug localization abilities. (2) FullStack-Learn, an innovative data-scaling and self-improving method that back-translates crawled and synthesized website repositories to improve the backbone LLM of FullStack-Dev. (3) FullStack-Bench, a comprehensive benchmark that systematically tests the frontend, backend and database functionalities of the generated website. Our FullStack-Dev outperforms the previous state-of-the-art method by 8.7%, 38.2%, and 15.9% on the frontend, backend, and database test cases respectively. Additionally, FullStack-Learn raises the performance of a 30B model by 9.7%, 9.5%, and 2.8% on the three sets of test cases through self-improvement, demonstrating the effectiveness of our approach. The code is released at https://github.com/mnluzimu/FullStack-Agent.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。