Skip to content

This repository is a starter kit for building "skills" and "agents" for Claude Code. The current content focuses on patterns, conventions, and agents for Java projects (Spring Boot, JUnit, LangChain4J), but the kit is designed to be extensible and multi-language (PHP, TypeScript, Python, etc.).

License

Notifications You must be signed in to change notification settings

giuseppe-trisciuoglio/developer-kit

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Developer Kit for Claude Code

A curated collection of reusable skills and agents for automating development tasks in Claude Code — focusing on Java/Spring Boot patterns with extensibility to PHP, TypeScript, and Python

Developer Kit for Claude Code teaches Claude how to perform development tasks in a repeatable way across multiple languages and frameworks. It includes specialized agents for code review, testing patterns, REST API design, and AI integration.

50 Total Skills — Comprehensive coverage of Spring Boot, testing, AI integration, cloud development, AI engineering patterns, and documentation generation

🚀 Quick Start

Claude Code CLI

/plugin marketplace add giuseppe-trisciuoglio/developer-kit

Claude Desktop

Enable Skills in Settings

🛠️ Installation & Setup

Claude Code CLI

# Install from marketplace
/plugin marketplace add giuseppe-trisciuoglio/developer-kit

# Or install from local directory
/plugin install /path/to/developer-kit

Claude Desktop

  1. Go to Settings > Capabilities
  2. Enable Skills toggle
  3. Browse available skills or upload custom skills
  4. Start using in conversations

🔌 Multi-CLI Support: Install in Other IDEs & Tools

The Developer Kit now supports installation across multiple AI-powered development environments through a unified Makefile interface.

Quick Start with Makefile

# Clone the repository
git clone https://github.com/giuseppe-trisciuoglio/developer-kit.git
cd developer-kit-claude-code

# See all available options
make help

# Install for your specific CLI tool
make install-copilot      # For GitHub Copilot CLI
make install-opencode     # For OpenCode CLI
make install-codex        # For Codex CLI
make install              # Auto-install for all detected CLIs

Supported CLI Tools

GitHub Copilot CLI

# Install agents for Copilot CLI
make install-copilot

# Installation creates:
# ~/.copilot/agents/          # Specialized agents for code review, testing, etc.

Features:

  • 14 Specialized Agents: Code review, architecture, security, testing, debugging experts
  • Usage: /agent to select agents or mention in prompts
  • Integration: Works with Copilot's native agent system

OpenCode CLI

# Install both agents and commands for OpenCode CLI
make install-opencode

# Installation creates:
# ~/.config/opencode/agent/     # Development agents
# ~/.config/opencode/command/  # Custom slash commands

Features:

  • 14 Development Agents: Full suite of specialized agents
  • 32 Custom Commands: From code generation to debugging and security reviews
  • Usage: @agent-name for agents, /command-name for commands
  • Discovery: Tab completion and command discovery

Codex CLI

# Install for Codex CLI
make install-codex

# Installation creates:
# ~/.codex/instructions.md    # Agent context and usage guide
# ~/.codex/prompts/           # Custom prompt commands

Features:

  • 32 Custom Prompts: Specialized commands for development workflows
  • 14 Agents Documentation: Complete agent descriptions and usage
  • Usage: /prompts:<name> to invoke custom commands

Management Commands

# Check installation status
make status

# Create backup before installing
make backup

# Remove all Developer Kit installations
make uninstall

# List available components
make list-agents      # Show all 14 agents
make list-commands    # Show all 32 commands
make list-skills      # Show all 50 skills by category

Installation Safety

  • Automatic Backups: Creates timestamped backups before installation
  • Conflict Resolution: Preserves existing configurations
  • Rollback Support: Easy uninstall to restore previous state
  • Version Tracking: Tracks what's installed from this kit

🎯 Local Project Installation

Install skills, agents, and commands directly into your local project for team-based development:

Interactive Claude Code Installation

# Clone the repository
git clone https://github.com/giuseppe-trisciuoglio/developer-kit.git
cd developer-kit-claude-code

# Run interactive installer for Claude Code
make install-claude

Interactive Features:

  • Environment Validation: Confirms Claude Code usage
  • 🎯 Category Selection: Choose specific skill categories
  • 🔧 Custom Selection: Pick specific agents and commands
  • 🛡️ Conflict Handling: Decide how to handle existing files
  • 📊 Progress Tracking: Real-time installation progress
  • 📋 Summary Report: Complete installation summary

Example Installation Flow

$ make install-claude

━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
      Claude Code Interactive Developer Kit Installer
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━

⚠ This installer is designed for Claude Code only.

Are you installing for Claude Code? (y/N): y

Step 1: Target Project
Enter the project path (absolute or relative): ~/my-spring-project

Step 2: Select Skill Categories
Available skill categories:
  1) AWS Java Skills (10 skills)
  2) AI Skills (3 skills)
  3) JUnit Test Skills (15 skills)
  4) LangChain4j Skills (8 skills)
  5) Spring Boot Skills (13 skills)
  6) Spring AI Skills (1 skill)
  7) All Skills
  8) None (skip skills)

Select categories (comma-separated, e.g., 1,4,5): 4,5

Step 3: Select Agents
  1) All Agents (14 available)
  2) Select specific agents
  3) None (skip agents)
Choose option [1-3]: 2

Available agents:
   1) java-documentation-specialist
   2) java-refactor-expert
   3) java-security-expert
   ...
Select agents (comma-separated numbers, or type 'all'): 1,3

Step 4: Select Commands
  1) All Commands (32 available)
  2) Select specific commands
  3) None (skip commands)
Choose option [1-3]: 1

━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
Starting Installation...
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━

Installing Skills...
  Category: LangChain4j Skills
    ✓ Installed: langchain4j-ai-services-patterns
    ✓ Installed: langchain4j-mcp-server-patterns
  Category: Spring Boot Skills
    ✓ Installed: spring-boot-actuator
    ✓ Installed: spring-boot-cache
    ...

Installing Selected Agents...
  ✓ Installed: java-documentation-specialist.md
  ✓ Installed: java-security-expert.md

Installing All Commands...
  ✓ Installed: devkit.java.code-review.md
  ✓ Installed: devkit.java.write-unit-tests.md
  ...

━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
✓ Installation Complete!
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━

Target directory: /Users/you/my-spring-project/.claude/
Files installed:  52
Files skipped:    0

Next Steps:
  1. Navigate to your project: cd /Users/you/my-spring-project
  2. Start Claude Code in the project directory
  3. Your skills, agents, and commands are now available!

What Gets Installed

my-project/
├── .claude/
│   ├── skills/                      # Auto-discovered skills
│   │   ├── langchain4j-ai-services-patterns/
│   │   ├── spring-boot-actuator/
│   │   └── ... (selected skills)
│   ├── agents/                      # @agent-name access
│   │   ├── java-documentation-specialist.md
│   │   ├── java-security-expert.md
│   │   └── ... (selected agents)
│   └── commands/                    # /command-name access
│       ├── devkit.java.code-review.md
│       ├── devkit.java.write-unit-tests.md
│       └── ... (selected commands)

Team-Based Development

For Teams Sharing Projects:

  1. Install Once: Use make install-claude in the project root
  2. Git Integration: All .claude/ files are version-controlled
  3. Team Consistency: Everyone gets the same tools and patterns
  4. Custom Skills: Create project-specific skills shared with team

Benefits:

  • 🔄 Consistent Tooling: Team uses same agents, skills, commands
  • 📚 Project Context: Skills understand your specific project structure
  • 🎯 Domain-Specific: Tailored to your business domain and patterns
  • 🚀 Quick Onboarding: New team members get all tools immediately

📚 Available Skills

Spring Boot Skills

Skill Purpose Key Topics
spring-boot-actuator Production monitoring and metrics Health checks, custom endpoints, metrics, startup tracking
spring-boot-dependency-injection Constructor injection best practices DI patterns, testing strategies, anti-patterns
spring-boot-crud-patterns REST CRUD API design with DDD Feature-based architecture, CRUD generator, Lombok, Spring Data
spring-boot-event-driven-patterns Event-driven architecture Domain events, Kafka, Spring Cloud Stream, transactional patterns
spring-boot-rest-api-standards REST API design standards HTTP semantics, error handling, pagination, security headers
spring-boot-test-patterns Integration testing patterns Testcontainers, Spring slice tests, database strategies
spring-boot-cache Spring Boot caching patterns Cache configuration, eviction strategies, distributed caching
spring-boot-saga-pattern Distributed transaction management Saga pattern, choreography, orchestration, compensating transactions
spring-boot-resilience4j Fault tolerance patterns Circuit breaker, retry, rate limiting, bulkhead patterns
spring-boot-security-jwt JWT authentication & authorization Bearer tokens, OAuth2/OIDC, RBAC, permission-based access control
spring-data-jpa Spring Data JPA best practices Query methods, custom repositories, performance optimization
spring-data-neo4j Neo4j graph database integration Graph modeling, Cypher queries, relationships, reactive mode
spring-boot-openapi-documentation OpenAPI/Swagger documentation API documentation, schema generation, SpringDoc

JUnit Testing Skills

Skill Purpose Key Topics
unit-test-service-layer Service layer unit testing Mocking, assertions, test organization
unit-test-controller-layer Controller layer testing Mock MVC, request/response handling, status codes
unit-test-parameterized Parameterized tests JUnit 5 @ParameterizedTest, multiple scenarios
unit-test-exception-handler Exception handling tests Custom exceptions, error scenarios
unit-test-bean-validation Jakarta Validation testing Bean validation constraints, custom validators
unit-test-security-authorization Spring Security testing @WithMockUser, authorization checks, roles
unit-test-application-events Application event testing Event publishing, listeners, @EventListener
unit-test-scheduled-async Async and scheduled task testing @Async, @Scheduled, TestScheduler
unit-test-json-serialization JSON serialization testing Jackson, @JsonTest, custom serializers
unit-test-config-properties Configuration properties testing @ConfigurationProperties, binding validation
unit-test-mapper-converter Mapper and converter testing Entity to DTO, custom converters
unit-test-caching Caching layer testing @Cacheable, cache eviction, hit/miss scenarios
unit-test-boundary-conditions Boundary condition testing Edge cases, limits, null values
unit-test-utility-methods Utility method testing Static methods, helpers, common functions
unit-test-wiremock-rest-api REST API mocking with WireMock HTTP stubs, response matching, status codes

LangChain4J Skills

Skill Purpose Key Topics
langchain4j-spring-boot-integration LangChain4J with Spring Boot Configuration, bean management, integration patterns
langchain4j-rag-implementation-patterns Retrieval-Augmented Generation Vector stores, retrievers, prompt chaining
langchain4j-ai-services-patterns AI service design Service patterns, model configuration, error handling
langchain4j-mcp-server-patterns Model Context Protocol servers MCP integration, tool exposure, data access
langchain4j-tool-function-calling-patterns Tool and function calling Agent patterns, tool definition, function orchestration
langchain4j-testing-strategies Testing LangChain4J applications Mock models, test containers, integration tests
langchain4j-vector-stores-configuration Vector store configuration Embeddings, similarity search, provider setup
qdrant-java-development Qdrant vector database integration Java, Spring Boot, Langchain4j, RAG, vector search

Spring AI Skills

Skill Purpose Key Topics
spring-ai-mcp-server-patterns Model Context Protocol servers MCP integration, Spring AI, tool exposure

AWS Java Skills

Skill Purpose Key Topics
aws-rds-spring-boot-integration Aurora RDS configuration with Spring JPA Cluster setup, connection pooling, read/write split, failover
aws-sdk-java-v2-core AWS SDK v2 core patterns Client configuration, credential management, async operations
aws-sdk-java-v2-s3 S3 operations with SDK v2 Bucket operations, object upload/download, presigned URLs
aws-sdk-java-v2-dynamodb DynamoDB operations Table operations, queries, scans, transactions
aws-sdk-java-v2-lambda Lambda function development Handler patterns, event processing, custom runtimes
aws-sdk-java-v2-secrets-manager Secrets Manager integration Secret retrieval, rotation, caching
aws-sdk-java-v2-kms KMS encryption operations Key management, encryption/decryption, envelope encryption
aws-sdk-java-v2-messaging SQS/SNS messaging patterns Queue operations, topic publishing, message processing
aws-sdk-java-v2-rds RDS management operations Instance management, snapshots, parameter groups
aws-sdk-java-v2-bedrock AWS Bedrock AI integration Model invocation, streaming, guardrails

AI Engineering Skills

Skill Purpose Key Topics
chunking-strategy Document chunking for AI systems Text splitting strategies, semantic chunking, RAG prep
prompt-engineering Advanced prompt design patterns CoT, few-shot, system prompts, optimization frameworks
rag Retrieval-Augmented Generation patterns Vector search, document processing, embedding strategies

🤖 Available Agents

Agent Specialization Best For
spring-boot-code-review-specialist Spring Boot code review Architecture, patterns, security review
spring-boot-code-review-expert Expert Spring Boot code review Advanced code analysis, security, performance review
spring-boot-backend-development-expert Backend development Implementation, testing, optimization
langchain4j-ai-development-expert AI integration LangChain4J implementation, RAG, agents
spring-boot-unit-testing-expert Unit testing Test patterns, Mockito, test organization
java-refactor-expert Java Refactoring Complex class refactoring, DDD alignment, legacy code modernization
java-software-architect-review Software architecture review High-level architecture, design patterns, scalability
java-security-expert Java security analysis Security vulnerabilities, OWASP, secure coding practices
java-documentation-specialist Java project documentation API docs, architecture guides, technical documentation generation
java-tutorial-engineer Java/Spring Boot tutorials Educational content, hands-on learning, step-by-step guides
prompt-engineering-expert Prompt engineering LLM prompt optimization, CoT, few-shot patterns
general-debugger Debugging and root cause analysis Error troubleshooting, test failures, performance issues

✨ Key Features

Specialized — Domain-specific agents for code review, testing, and AI development tailored to Java/Spring Boot

Composable — Skills stack together automatically. Claude identifies which skills are needed and uses them in combination

Portable — Use the same skills across Claude.ai, Claude Code CLI, Claude Desktop, and the Claude API

Efficient — Skills load on-demand, consuming minimal tokens until they're actively used

🏗️ Architecture & Patterns

This kit promotes a feature-based, DDD-inspired architecture for Java/Spring Boot projects:

feature/
├── domain/              # Domain-pure logic
│   ├── model/           # Spring-free entities
│   ├── repository/      # Domain ports (interfaces)
│   └── service/         # Domain services
├── application/         # Use cases & business logic
│   └── service/         # Application services (@Service)
├── presentation/        # HTTP layer
│   ├─ rest/            # REST controllers & mappers
│   └── dto/             # Immutable DTOs/records
└── infrastructure/      # Technical adapters
    └── persistence/     # JPA repositories & adapters

Core Principles

  • ✅ Constructor injection exclusively (never field injection)
  • ✅ Java records (16+) preferred for DTOs
  • ✅ Repository pattern with domain interfaces
  • ✅ Immutability throughout (final fields, @Value, records)
  • ✅ Clean separation: domain ↔ framework

🔧 Technology Stack

Category Technology Version
Java OpenJDK/GraalVM 16+
Framework Spring Boot 3.x
Testing JUnit 5 (Jupiter) + Mockito + AssertJ 5.x / 4.x+
Integration Testing Testcontainers 1.19.0+
Data Spring Data JPA Jakarta Persistence
Graph Database Spring Data Neo4j 7.x
Validation Jakarta Validation 3.0+
Utilities Lombok 1.18.30+
Messaging Spring Cloud Stream 4.x
AWS SDK AWS SDK for Java v2 2.20+
Cloud AWS (RDS Aurora, S3, Lambda, DynamoDB) Latest
AI/ML LangChain4J, AWS Bedrock Latest

🎯 Use Cases

Code Review & Architecture — Automated Spring Boot code reviews, architecture validation, security analysis

REST API Design — Standardized API development with proper HTTP semantics, error handling, pagination

Testing Strategies — Unit test patterns, integration testing with Testcontainers, test organization

AI Integration — LangChain4J implementation guidance, RAG patterns, agent development, MCP server creation

Event-Driven Architecture — Domain events, event sourcing patterns, Kafka integration, transactional consistency

AWS Cloud Development — RDS Aurora configuration, S3 operations, Lambda functions, DynamoDB, Secrets Manager, KMS encryption

📖 How to Use

For Individual Contributors

  1. Install the skill/agent in Claude Code
  2. Describe your development task
  3. Claude will automatically load relevant skills and provide guidance

For Teams

  1. Clone or fork this repository
  2. Customize skills for your team's standards
  3. Deploy via /plugin install /path/to/developer-kit-claude-code
  4. Share the plugin marketplace link with your team

🔧 Custom Commands

This kit includes specialized commands for workflow orchestration, verification, skill validation, and Java development automation:

Project Workflow Commands

Command Purpose Use Case
/devkit.generate-changelog Generate project changelog Create and maintain CHANGELOG.md for any project type with Git integration
/devkit.write-a-minute-of-a-meeting Generate meeting minutes from transcripts Create professional meeting summaries and action items from meeting transcripts or notes
/devkit.prompt-optimize Optimize prompts for better AI performance Enhance prompt engineering for improved Claude responses and task execution

GitHub Commands

Command Purpose Use Case
/devkit.github.create-pr Create GitHub pull request Create branch, commit changes, and submit PR with automated description and commit messages
/devkit.github.review-pr Comprehensive GitHub PR review Perform code quality, security, performance, architecture, and testing review of a PR

Security Commands

Command Purpose Use Case
/devkit.java.security-review Java enterprise security audit Comprehensive security review for Spring Boot and Jakarta EE applications (OWASP, CVEs, configs)
/devkit.ts.security-review TypeScript/Node.js security audit Security review for TypeScript applications (Next.js, NestJS, Express) with OWASP Top 10 analysis
/devkit.generate-security-assessment Generate security assessment document Create comprehensive security assessment documentation after security audits (English by default)

GitHub Spec Kit Commands

Command Purpose Use Case
/speckit.check-integration Verify task integration with existing codebase Run AFTER /speckit.tasks to detect duplication and integration opportunities
/speckit.optimize Optimize execution plan for parallelization Run AFTER check-integration to plan subagent delegation and resource allocation
/speckit.verify Comprehensive implementation verification Run AFTER /speckit.implement to validate all requirements, tests, and code quality

Java Development Commands

Command Purpose Use Case
/devkit.java.generate-crud Generate CRUD implementation for domain class Create complete REST API with controllers, services, DTOs using spring-boot-crud-patterns
/devkit.java.generate-docs Generate Java project documentation Create comprehensive API docs, architecture guides, Javadoc, and technical documentation
/devkit.java.write-unit-tests Generate comprehensive JUnit 5 unit tests Create unit tests with Mockito, AssertJ for Java classes
/devkit.java.write-integration-tests Generate Spring Boot integration tests Create Testcontainers-based tests for complete workflow testing
/devkit.java.code-review Perform comprehensive Java code review Review code quality, architecture, security, and performance
/devkit.java.security-review Security-focused code review Identify vulnerabilities, OWASP compliance, secure coding practices
/devkit.java.architect-review High-level architecture review Validate design patterns, scalability, and architectural decisions
/devkit.java.refactor-class Intelligent refactoring assistant Refactor complex Java classes with Clean Architecture, DDD patterns, and Spring Boot best practices
/devkit.generate-refactoring-tasks Generate step-by-step refactoring plan Create detailed, actionable refactoring tasks for complex Java classes
/devkit.java.dependency-audit Comprehensive dependency audit Scan vulnerabilities, verify licenses, and get update recommendations
/devkit.java.upgrade-dependencies Safe dependency upgrade strategies Upgrade dependencies with compatibility testing and rollback procedures

Project Workflow Commands

Command Purpose Use Case
/devkit.generate-changelog Generate project changelog Create and maintain CHANGELOG.md for any project type with multi-language support
/devkit.write-a-minute-of-a-meeting Generate meeting minutes from transcripts Create professional meeting summaries and action items from meeting transcripts or notes
/devkit.prompt-optimize Optimize prompts for AI performance Enhance prompt engineering and save optimized prompt to optimized-prompt.md

Skill Development Commands

Command Purpose Use Case
/devkit.verify-skill Validate skill against DevKit standards Verify a skill's compliance with requirements, format, and best practices

Long-Running Agent (LRA) Commands

Commands for managing complex projects that span multiple context windows, based on Anthropic's research on long-running agents.

Command Purpose Use Case
/devkit.lra.init Initialize LRA environment Set up feature list, progress tracking, and init script for multi-session projects
/devkit.lra.start-session Start a new coding session Read progress, check health, select next feature at the beginning of each session
/devkit.lra.add-feature Add feature to the list Add new requirements discovered during development
/devkit.lra.mark-feature Mark feature as passed/failed Update feature status after implementation and testing
/devkit.lra.checkpoint Create session checkpoint Commit changes, update progress log, ensure clean state at session end
/devkit.lra.status View project status Display progress metrics, priorities, and recent activity
/devkit.lra.recover Recover from broken state Diagnose issues, revert if needed, restore working state

LRA Workflow:

/devkit.lra.init [project-description]     ← First time setup
    ↓
/devkit.lra.start-session                  ← Start of every session
    ↓
... implement ONE feature ...
    ↓
/devkit.lra.mark-feature [id] passed       ← After testing
    ↓
/devkit.lra.checkpoint [summary]           ← End of every session

📖 Complete LRA Workflow Guide — Detailed documentation with examples and best practices.

GitHub Spec Kit Workflow

/speckit.tasks
    ↓
/speckit.check-integration    ← Detect existing code & integration opportunities
    ↓
/speckit.optimize             ← Plan parallelization & subagent assignment
    ↓
/speckit.implement            ← Execute with optimization
    ↓
/speckit.verify               ← Verify completeness & quality

Spec Kit Command Details

/speckit.check-integration

  • Analyzes tasks.md for codebase integration opportunities
  • Detects duplication risks and existing implementations
  • Verifies architectural alignment
  • Output: Integration analysis report (READ-ONLY)
  • Timing: After task generation, before optimization

/speckit.optimize

  • Builds dependency graph and identifies parallelization opportunities
  • Allocates tasks to specialized subagents
  • Generates phased execution plan with validation checkpoints
  • Output: Optimization report with phase execution matrix (READ-ONLY)
  • Timing: After integration check, before implementation

/speckit.verify

  • Validates all tasks completed
  • Confirms requirements coverage
  • Runs test suites and checks coverage metrics
  • Audits code quality, security, and performance
  • Generates verification-report.md
  • Output: verification-report.md (PASS/FAIL)
  • Timing: After implementation execution

Developer Kit Command Details

/devkit.verify-skill

  • Validates skill compliance with DevKit standards and official Skills documentation
  • Checks SKILL.md frontmatter (name, description, allowed-tools, etc.)
  • Verifies file structure and syntax (no YAML errors, valid paths)
  • Validates content quality (clear triggers, instructions, examples)
  • Confirms all referenced files exist in skill directory
  • Usage: /devkit.verify-skill skill-name
  • Output: Comprehensive validation report with PASS/FAIL status
  • Best for: Skill authors before submitting PRs or publishing

/devkit.java.refactor-class

  • Intelligent refactoring assistant for complex Java classes
  • Applies Clean Architecture, DDD patterns, and Spring Boot best practices
  • Supports multiple refactoring scopes: cleanup, architecture, performance, security, testing, comprehensive
  • Includes safety checks, backup strategies, and incremental refactoring process
  • Usage: /devkit.java.refactor-class [class-file-path] [refactoring-scope] [options]
  • Scopes: cleanup, architecture, performance, security, testing, comprehensive
  • Options: dry-run, backup, validate-only
  • Output: Refactored code with detailed summary of changes and quality metrics
  • Best for: Improving code quality, maintainability, and architectural alignment

🌍 Language Roadmap

  • ✅ Java/Spring Boot — Comprehensive patterns and agents
  • 📋 TypeScript — Node.js, NestJS, Express patterns (planned)
  • 📋 Python — Django, FastAPI patterns (planned)
  • 📋 PHP — Laravel, Symfony patterns (planned)

🤝 Contributing

Adding a New Skill

  1. Create skills/<domain>/<skill-name>/SKILL.md
    • For Spring Boot: skills/spring-boot/spring-boot-*/SKILL.md
    • For JUnit Testing: skills/junit-test/unit-test-*/SKILL.md
    • For LangChain4J: skills/langchain4j/langchain4j-*/SKILL.md
    • For AWS Java: skills/aws-java/aws-*/SKILL.md
  2. Include YAML frontmatter (name, description, language)
  3. Add practical examples and best practices
  4. Document dependencies and commands
  5. Add supporting files (examples.md, references.md) if needed
  6. Update marketplace.json with new skill path
  7. Submit a PR with description and examples

Adding a New Domain

  1. Create skills/<new-domain>/ directory
  2. Add domain-specific skills following the pattern
  3. Create corresponding agent if needed
  4. Follow existing structure for consistency
  5. Update README.md with new domain section
  6. Update marketplace.json with new paths

See CONTRIBUTING guidelines for detailed instructions.

🔒 Security & Best Practices

⚠️ Important: Skills can execute code in Claude's environment. Review all custom skills before deploying.

Best Practices

  • Only install skills from trusted sources
  • Review SKILL.md and all scripts before enabling
  • Keep skills version-controlled with git
  • Audit regularly for updates and changes
  • Test in non-production environments first

📘 Complete Guides

Comprehensive documentation for all Developer Kit components:

Skills Guides

  • Spring Boot Skills Guide - Complete guide to Spring Boot patterns including dependency injection, actuator, caching, data persistence, REST APIs, resilience patterns, and testing strategies
  • JUnit Testing Skills Guide - Comprehensive testing patterns for controller, service, data, external integrations, infrastructure, and advanced testing scenarios
  • LangChain4j Skills Guide - Building AI-powered applications with core integration, AI services, RAG implementation, vector stores, and MCP server patterns
  • AWS Java SDK Skills Guide - Cloud development with AWS services including core SDK setup, storage (S3), databases (DynamoDB, RDS), compute (Lambda), messaging (SQS/SNS), and security services

Component Guides

  • Agents Guide - Complete documentation for specialized agents including Java development, testing & quality, AI & LangChain4j, documentation, and engineering agents
  • Commands Guide - Comprehensive command reference for Java/Spring Boot development, testing, DevKit management, and utility commands with practical workflows
  • LRA Workflow Guide - Long-Running Agent workflow for multi-session projects based on Anthropic's research on effective agent harnesses

📚 Resources

Official Documentation

Related Projects

📝 License

This project is licensed under the LICENSE file in the repository root. See that file for details.

📞 Support

  • Questions? Open an issue
  • Contributions? Submit a pull request
  • Integration proposals? Create a discussion or reach out

📅 Changelog

For a complete history of changes, please see the dedicated CHANGELOG.md file.

Recent highlights:

  • v1.15.0:
    • New comprehensive Makefile with multi-CLI support for GitHub Copilot CLI, OpenCode CLI, and Codex CLI
    • Interactive Claude Code installer (make install-claude) with category-based skill selection, conflict handling, and project-specific installation
    • Support for team-based development with version-controlled .claude/ configurations
    • Enhanced installation management with backup, status, and uninstall capabilities
  • v1.14.0:
    • New spring-boot-security-jwt skill with JWT authentication patterns, OAuth2/OIDC integration, and permission-based access control
    • New spring-ai-mcp-server-patterns for Model Context Protocol integration
    • Enhanced /devkit.feature-development command with AskUserQuestion tool and improved agent fallback order
  • v1.13.0: New /devkit.feature-development command with systematic 9-phase approach for guided feature development, including three new general-purpose agents (explorer, architect, reviewer) with Task tool integration and fallback mechanism
  • v1.12.1: Fixed commands for correct agent selection on execution
  • v1.12.0: Long-Running Agent (LRA) workflow commands and guide for multi-session project management
  • v1.11.0: Enhanced development workflow with new commands, agents, and comprehensive documentation updates
  • v1.10.0: GitHub integration and workflow automation with PR creation and review commands
  • v1.9.0: Documentation generation commands and educational content expansion

Made with ❤️ for Java/Spring Boot developers using Claude Code

About

This repository is a starter kit for building "skills" and "agents" for Claude Code. The current content focuses on patterns, conventions, and agents for Java projects (Spring Boot, JUnit, LangChain4J), but the kit is designed to be extensible and multi-language (PHP, TypeScript, Python, etc.).

Topics

Resources

License

Contributing

Stars

Watchers

Forks

Packages

No packages published