Open-source WaveMaker alternatives

JeecgBoot is an enterprise-grade low-code development platform for Java projects, with low-code tooling, an AI application platform and an AI knowledge base at its core. Its documentation is mainly in Chinese, with English and Japanese versions available. Version 2.0 of its low-code approach lets developers describe a system in natural language and have AI generate the front-end and back-end code, database tables and menu permissions.

The platform includes a code generator that handles single-table, tree, one-to-one and one-to-many data models, along with online forms, a form designer, a workflow designer, a portal designer, report and big-screen designers and OA office features. Its AI layer provides a chat assistant, a knowledge base, flow orchestration, MCP and plugins, and works with mainstream models such as ChatGPT, DeepSeek and Ollama. Fine-grained permissions cover buttons and data.

The stack pairs a Vue 3, TypeScript, Vite and Ant Design Vue front end with a Spring Boot, Spring Cloud Alibaba and MyBatis-Plus back end. JeecgBoot is Apache-2.0 licensed and aimed at Java teams that want to cut repetitive CRUD work while keeping source-level control.

Oinone is a low-code framework positioned as infrastructure for AI-native enterprise applications. It is fully metadata-driven: a shared metadata model is the common ground for AI and developers, so applications stay maintainable and can evolve. This repository, Oinone Pamirs, holds the Java back end. It is released under the AGPL-3.0 license, and the project uses the slogan about AI giving speed and Oinone giving rigor.

The project promotes AI coding and so-called vibe coding on top of a structured model rather than free-form generated code. A one-command quickstart using Docker brings up a backend, and the first boot takes a few minutes. Documentation is available through a website, with a Chinese mirror on Gitee.

Oinone is aimed at Java teams building internal systems and enterprise software, such as business management applications, who want low-code productivity without giving up control of the code. Treat the README's claims about AI-native development as vendor positioning, and check the repository for licensing and edition details.

Structr is a generic low-code development and runtime platform built on graph technology. It is not tied to one use case and is meant for building fully custom applications, including web applications, graph-database-backed back ends, data-centric systems, middleware and integration components.

At its core Structr keeps both application components, such as schema, UI definitions, templates, scripts and workflows, and the application data in a graph database (Neo4j). That gives schema flexibility and expressive data modeling without join-heavy queries, and the schema can be represented as JSON Schema with import and export. A virtual file system manages files with custom metadata and can store binary content on external providers such as S3. Development happens in a web-based developer and admin UI that works as an integrated IDE.

The README is explicit about what Structr is not: it is not a prebuilt CMS, CRM or ERP, a website builder, or an industry template collection, but a set of building blocks for your own applications. The GitHub repository is a clone, and development and issue tracking have moved to gitlab.structr.com. The license is listed as Other, and the topics mention GraalVM polyglot programming.

Skyve is an open-source low-code platform for building secure, scalable enterprise applications. Rather than generating large amounts of code, applications are described in concise XML metadata, and the platform interprets those declarations at runtime to enforce security, render user interfaces, manage persistence and evolve the database schema.

Security is declarative: permissions are set per document with role-based create, read, update and delete rules and row-level scoping, and are enforced across persistence, queries and the UI. REST and UI responses only include fields a user may see, unpermitted input is ignored, access paths are computed from roles, and REST access is blocked unless it has been explicitly allowed. The whole metadata model is validated statically on every build.

This repository is the Java implementation of the Skyve framework specification, with Maven commands for creating, building and updating projects and a tutorial, user guide, development guide and cookbook. The project presents the metadata approach as well suited to AI-assisted development, and it is licensed under LGPL-2.1.

Joget is an application platform that combines rapid application development, business process automation and workflow management in a visual, web-based builder. Business and technical teams can use it together to build full enterprise applications without much hand coding, aiming to shorten delivery from months to weeks or days. Apps built on it are mobile ready and cloud ready, and the platform offers APIs for integration and a plugin architecture for extension.

The README describes a marketplace where ready-made apps, plugins and templates can be downloaded, along with a community Q&A forum, a knowledge base, video tutorials, an online academy and translations in more than twenty languages. Deployment is flexible: on-premise, on-demand cloud, Docker, Kubernetes, Cloud Foundry, OpenShift and several cloud marketplaces are listed.

Licensing is split. Joget Community Edition is released under GPL version 3, while the Professional, Enterprise and Large Enterprise editions are covered by a commercial end-user license agreement. The core is written in Java. It suits organizations that want to digitize processes such as approvals and forms with a platform they can run themselves.