Rules, Skills & Agents for Java

Stargazers over time

Goal

A curated collection of Rules, Skills, and Agents for Java Enterprise development, designed to streamline modern SDLC workflows.

The project add support for:

Agile: User Stories & Gherkin
Architecture: ADRs& UML / C4 / ER Diagrams`
AI Tooling: AGENTS.md & AI Planning
Java development: Build system based on Maven, Design, Coding, Testing, Observability, Refactoring & JMH Benchmarking, Performance testing with JMeter, Profiling with Async profiler/OpenJDK tools & Documentation
Java frameworks: Spring Boot, Quarkus & Micronaut with the same support (Core, Rest, Jdbc, ORM, Flyway, Unit Tests, IT & AT).

Deliverables

The project generates a set of deliverables at the end of any iteration.

Deliverable	Location
System prompts for Java	Catalog (`.cursor/rules`)
Skills for Java	https://skills.sh/jabrena/cursor-rules-java
Agents for Java	Catalog (`.cursor/agents`)

Compatibility

This project is compatible with any Tool compatible with Skills, Agents or AGENTS.md like Cursor AI, Claude Code & Github Copilot.

Getting started

Read the following comprehensive guides to use this project today.

How to use them?

The SDLC has evolved with the arrival of this new set of AI tooling, enhancing the Software Engineering process. In the development of this project, it was identified 3 different workflows: Prompting Enginering Workflow, Pipelines Workflow & Agentic Workflow.

Prompting Enginering Workflow

In this workflow, the Software engineer interact with models using User prompts and in an incremental way you delegate a delegate completely a task or ask help in certain moments. You could use this project to refactor the code generated or delegate the task and associate a System prompt / Skills to that task.

Pipelines Workflow

Adding AI tools to your pipeline can provide new opportunities to deliver more value (examples: automatic coding, code refactoring, continuous profiling, and others).

Agentic Workflow

Agents for Java Enterprise development were designed to cover the whole SDLC starting from the requirements and the architectural process. Once you have the foundations well designed. it is possible to execute Plans, iterate and discuss it and when the plans associated to User Stories has the right size & complexity, it is easy to pass to the models to be implemented with security in other case, you will have the classical issues with implementations without control.

Limitations

Lack of determinism

From the outset, be aware that the results provided by interactions with the different Cursor rules are not deterministic due to the nature of the models, but this can be mitigated with clear goals and validation checkpoints.

Limits of interactions with models

Models are able to generate code, but they cannot run code with your local data. To address this limitation, some prompts provide scripts to bridge this gap on the model side.

Contribute

If you have great ideas, read the following document to contribute.

Examples

The repository includes a collection of examples where you can explore the possibilities of these system prompts designed for Java.

Architectural decision records, ADR

Date	ID	Name
2026-03-01	ADR-004	Skill Generation
2025-09-16	ADR-003	Website Generation
2025-07-10	ADR-002	Cursor Rules scope configuration
2025-07-08	ADR-001	Cursor Rules generation from XML Files

Changelog

Review the CHANGELOG for further details

Java JEPS from Java 8

Java uses JEPs as the vehicle to describe new features to be added to the language. The repository continuously reviews which JEPs could improve any of the cursor rules present in this repository.

JEPS List

Meetups, Conferences, Workshops & Articles

cursor-rules-java