PJ Core Dev Tools - AI Coding Rules

This repository contains AI coding rules and patterns for PJ and Core development workflows. The CLI features have been moved to the nucli project.

📁 Structure

.cursor/
└── rules/                       # AI coding rules and patterns
    ├── clojure/                 # Clojure language rules
    │   ├── datomic/            # Datomic database rules
    │   ├── diplomat_architecture/  # Diplomat Architecture patterns
    │   └── testing/            # Testing guidelines
    ├── scala/                   # Scala language rules
    ├── pj/                      # PJ-specific rules and patterns
    ├── core/                    # Core-specific rules and patterns
    └── agents/                  # AI agent definitions (MD files)

🚀 Using the Rules

The rules in this repository are managed by the pj-core-dev-tools CLI, which is now part of the nucli project.

Quick Install

# The CLI is now in nucli - install from there
cd ~/dev/nu/nucli/nucli.d/pj-core-dev-tools.d
./scripts/install

# Then use it to install rules to your projects
pj-core-dev-tools rules install ~/dev/my-project

Available Rules

🔧 Clojure Rules

• Core Language: Idiomatic Clojure development patterns, data structures, and functional programming
• Diplomat Architecture: Complete architectural patterns for layered service design
• Datomic: Database interaction patterns and best practices
• Error Handling: Robust error handling and validation patterns
• Testing: Comprehensive testing strategies and patterns
• State Management: Concurrent state management patterns
• Performance: Java interop and performance optimization

📊 Scala Rules

• Dataset Patterns: Function-oriented dataset generation following data engineering best practices
• Unit Testing: Comprehensive unit testing patterns

🏢 PJ & Core Specific Rules

• PJ Rules: Business logic patterns specific to PJ (Pessoa Jurídica) domain
• Core Rules: Core banking patterns and best practices
• AI Agents: Individual AI agent definitions for specific use cases

🎯 Rule Categories

All rules are organized by:

• Language (clojure, scala)
• Domain (PJ, Core, architect patterns, testing)
• Specificity (core language features vs specialized frameworks)

📖 Adding AI Agents

AI agents are defined as individual .mdc files in the .cursor/rules/agents/ directory. Each agent file should:

1. Follow the .mdc format with frontmatter metadata
2. Include clear description of the agent's purpose
3. Define scope using globs patterns
4. Specify if always applied using alwaysApply: true/false

Example Agent Structure

---
description: "Agent description and purpose"
globs: ["**/*.clj", "**/pj/**"]
alwaysApply: false
---

# Agent Name

Agent-specific rules and guidelines...

Recommended Location for New Agents

Place new AI agent files in:

• .cursor/rules/agents/ - For general-purpose agents
• .cursor/rules/pj/agents/ - For PJ-specific agents
• .cursor/rules/core/agents/ - For Core-specific agents

🔧 Development

Adding New Rules

1. Create new .mdc files in the appropriate category folder
2. Follow the existing naming conventions and structure
3. Include comprehensive documentation and examples
4. Test with the pj-core-dev-tools CLI (from nucli)

Rule Format

Rules use .mdc (Markdown with Cursor extensions) format optimized for AI coding assistants:

• Clear, actionable patterns
• Comprehensive examples
• Business context and reasoning
• Anti-patterns to avoid
• Testing strategies

🤝 Contributing

1. Rules should be specific and actionable
2. Include business context and reasoning
3. Provide positive and negative examples
4. Follow existing naming conventions
5. Test rules with real projects before submitting

📜 License

This project is part of the Nu PJ and Core development ecosystem. See repository license for details.

---

Note: The CLI features for managing these rules have been moved to the nucli project for centralized tooling management. This repository now focuses exclusively on the AI coding rules and patterns.