JeecgBoot

JeecgBoot is an open‑source AI‑enhanced low‑code platform built for enterprise application development. The website calls it a “企业级 AI 低代码开发平台” that can generate an entire system's front‑end and back‑end code from a single natural‑language sentence. It is positioned as a way to create digital tools such as OA, ERP, CRM, MIS, industry‑specific management systems, data‑report dashboards, mobile apps, and AI knowledge bases.

The platform blends low‑code, zero‑code, and AI‑assisted development within a front‑back separation architecture. In practice, you type a description of a workflow, a form layout, or a report requirement into the AI prompt box. The AI interprets the sentence, creates an online configuration (for example a JSON or XML definition of the form fields, workflow nodes, or chart components), and then the built‑in code generator translates that configuration into Java source files for both the UI and the server side. After generation you can edit the configuration directly in the web UI, merge the generated code with existing modules, and later ask the AI to modify the code if requirements change. This “Skills 生成 → 在线配置 → 代码生成 → 手工合并 → AI修改” loop is intended to replace the repetitive work that typically accounts for about 90 % of Java project effort.

JeecgBoot's core modules include a code generator, an online form builder, a BPM workflow engine, a dashboard and large‑screen reporting tool, mobile‑app scaffolding, permission management, AI application orchestration, and an enterprise knowledge‑base. Each module works through a visual editor: the form builder lets you drag fields and define validation rules, the BPM engine lets you connect nodes to model approval processes, and the reporting tool lets you select data sources and chart types to produce a dashboard that can be exported as a web page. The output of these editors is a set of Java classes, HTML/JS assets, and configuration files that can be compiled and deployed as a standalone application.

The platform is model‑agnostic, meaning you can connect to a variety of large language models, DeepSeek, Tongyi Qianwen, Zhipu AI, OpenAI, Claude, Gemini, Ollama, whether hosted in the cloud or run locally. Integration is done by configuring API keys and endpoint URLs in the AI orchestration settings, after which the same natural‑language prompt workflow is available.