Development AI Agent Team Boilerplate

🚀 Next-Generation LLM CLI Runtime Development System Orchestrate specialized AI experts with automated project detection, intelligent collaboration, and zero-configuration deployment

Welcome to the Development AI Agent Team Boilerplate, an advanced project orchestration system designed for modern LLM CLI runtimes including Claude Code, Gemini CLI, Codex, and Junie. This repository provides a complete AI-powered development ecosystem with intelligent project detection, expert team assembly, and comprehensive quality management—all optimized for seamless LLM execution.

🌟 What Makes This Special?

This isn't just another boilerplate—it's a complete AI development orchestration platform that transforms how LLM CLI runtimes handle complex development projects. Built with extensive integration testing and real-world optimization, it delivers enterprise-grade development workflows through simple user interactions.

⚡ Revolutionary Features

🎯 Intelligent Project Detection: Advanced keyword scoring algorithm automatically identifies project types and assembles optimal expert teams
🧠 Integrated Memory System: Project-specific memory templates with real-time state preservation and context continuity across sessions
🏆 Quality-First Architecture: Automated quality gates with validation commands, performance targets, and continuous improvement tracking
🤝 Dynamic Expert Collaboration: Signal-based coordination system enabling seamless cross-expert consultation and task handoffs
🔄 Session Continuity: Advanced auto-compact handling with complete project state restoration
⚙️ Zero Configuration: Intelligent defaults with automatic system setup—just describe your project and start building

🏗️ Core Architecture Components

Component	Description	Integration Level
🎯 project-detection.yaml	Advanced project type detection with keyword scoring (40pts keywords + 35pts frameworks + 25pts technologies)	⚡ Auto-Detection
👥 team.yaml	9 specialized experts with 45+ granular capabilities and cross-domain collaboration matrix	🤝 Expert System
🧠 memory-templates.yaml	Project-specific memory structures (.memory/ folder) with automatic context preservation	💾 Memory System
🏆 quality-standards.json	Comprehensive quality frameworks with validation commands and automated measurement	✅ Quality Gates
🔄 orchestration.yaml	Complete development lifecycle with phased execution protocols and expert coordination	🎼 Workflow Engine
📋 user_req.yaml	Intelligent requirements capture with project constraint normalization	📝 Requirements
🤖 prompts/	9 expert prompt files with auto-loading validation and collaboration protocols	🚀 Expert Activation
💾 .memory/	Real-time project state management with session continuity and decision tracking	🔄 State Management

🚀 LLM CLI Runtime Support

LLM Runtime	Main Guidelines	Optimization Focus	Key Capabilities
🟦 Claude Code	`CLAUDE.md`	Advanced reasoning & memory integration	Auto-compact handling, session continuity, expert collaboration
🟢 Codex	`AGENTS.md`	Systematic analysis patterns	Dual PM system, enhanced coordination protocols
🟡 Gemini CLI	`GEMINI.md`	Creative multimodal problem-solving	Innovative approaches, visual integration, dynamic adaptation
🟣 Junie	`guidelines.md`	Ultra-efficient token optimization	JetBrains integration, ~20k token limit optimization

🎯 Expert Domain Specializations

Expert Domain	Core Expertise	Technology Focus	Integration Points
🎨 Frontend	Next.js architecture, component systems, UX optimization	React, TypeScript, Tailwind CSS	UI/API integration, performance optimization
⚙️ Backend	API design, database optimization, microservices	Nest.js, Python, authentication, scaling	Data architecture, security implementation
🔧 System Dev	AI/ML systems, video processing, GPU computing	TensorFlow, OpenCV, streaming, 3D conversion	High-performance computing, real-time systems
🌐 Full Stack	System architecture, cross-platform integration	End-to-end coordination, technology unification	Architecture decisions, integration orchestration
🚀 DevOps	Containerization, CI/CD, cloud deployment	Docker, automation, infrastructure as code	Deployment pipelines, monitoring, scaling
✅ QA	Automated testing, performance validation	Playwright MCP, accessibility, security testing	Quality gates, compliance verification
🔍 Research	Market analysis, technology evaluation	Strategic insights, competitive analysis	Technology recommendations, risk assessment
🛠️ MCP Tools	Tool orchestration, workflow automation	GitHub integration, Context7, Sequential Thinking	Advanced tool combinations, performance optimization

📁 Complete System Architecture

dev-ai-agent-team-boilerplate/
├── 🎯 Core Configuration
│   ├── project-detection.yaml      # Auto project type detection & expert assembly
│   ├── team.yaml                   # Expert definitions & collaboration matrix
│   ├── memory-templates.yaml       # Project-specific memory structures
│   ├── quality-standards.json      # Quality gates & validation frameworks
│   ├── orchestration.yaml          # Development lifecycle & task workflows
│   └── user_req.yaml              # Requirements capture template
├── 🚀 LLM Runtime Guidelines
│   ├── CLAUDE.md                   # Claude Code optimized guidelines
│   ├── AGENTS.md                   # Codex systematic analysis patterns
│   ├── GEMINI.md                   # Gemini multimodal capabilities
│   └── guidelines.md               # Junie token optimization
├── 🤖 Expert Activation System
│   └── prompts/
│       ├── pm.prompt.md            # Project management & coordination
│       ├── frontend.prompt.md      # Next.js & UI development
│       ├── backend.prompt.md       # API & database architecture
│       ├── fullstack.prompt.md     # System integration & architecture
│       ├── system-dev.prompt.md    # AI/ML & high-performance systems
│       ├── devops.prompt.md        # Deployment & infrastructure
│       ├── qa-testing.prompt.md    # Quality assurance & testing
│       ├── research.prompt.md      # Market analysis & tech research
│       └── mcp-tools.prompt.md     # Tool orchestration & automation
├── 💾 Memory Management System
│   └── .memory/
│       ├── active-context.md       # Real-time project status
│       ├── decisions.md            # Technical decision history
│       ├── collaboration.log.md    # Expert interaction tracking
│       ├── project-state.json      # Comprehensive project metrics
│       ├── session-history.json    # Session continuity data
│       └── artifacts.manifest.json # Generated deliverables catalog
├── 🛠️ Environment Configurations
│   ├── .claude/settings.json       # Claude auto-compact hooks
│   ├── .junie/                     # Junie-specific optimization files
│   └── .mcp/                       # MCP server configurations
└── 📚 Domain Expertise Guides
    ├── frontend.md                 # Next.js & React best practices
    ├── backend.md                  # API & database architecture
    ├── system-dev.md               # AI/ML & system development
    ├── fullstack.md                # Integration & architecture
    ├── deployment.md               # DevOps & containerization
    ├── qa-testing.md               # Testing & quality assurance
    ├── mcp-tools.md                # Tool orchestration guide
    ├── project-management.md       # PM protocols & coordination
    └── research.md                 # Strategic analysis methods

🚀 Getting Started - Experience Next-Level AI Development

🎯 Simple Project Creation

Before (Traditional approach):

"Create a web app with React, set up backend with Node.js, configure database, set up testing..."

Now (AI Agent Team approach):

"Create a todo management web app"
→ Auto-detects: Web Application Project
→ Assembles: Frontend + Backend + Full Stack + QA + DevOps experts
→ Starts: Complete development workflow automatically

📦 Installation Options

Choose your preferred LLM CLI runtime and get started instantly:

LLM Runtime	Installation Command	Auto-Setup Features
🟦 Claude Code	`./install-claude.sh`	Auto-compact hooks, session continuity, memory integration
🟢 Codex	`./install-codex.sh`	MCP configurations, systematic analysis setup
🟡 Gemini CLI	`./install-gemini.sh`	Multimodal configurations, creative problem-solving setup
🟣 Junie	`./install-junie.sh`	Token optimization, JetBrains integration

⚡ One-Command Installation

Universal Installation (Recommended)

# Clone and auto-setup
git clone https://github.com/HanTechnology/dev-ai-agent-team-boilerplate.git
cd dev-ai-agent-team-boilerplate

# Windows
install.bat

# Linux/macOS
chmod +x install.sh && ./install.sh

Direct LLM-Specific Installation

# For Claude Code (Recommended for advanced features)
./install-claude.sh

# For Codex (Systematic analysis focus)
./install-codex.sh

# For Gemini CLI (Creative multimodal approach)
./install-gemini.sh

# For Junie (Token-optimized for JetBrains)
./install-junie.sh

🎯 How It Works - Revolutionary AI Development

🚀 Intelligent Project Detection

Our advanced keyword scoring algorithm automatically identifies your project type:

Project Detection Algorithm:
- Keywords (40 points): "web app", "dashboard", "API", "mobile", "AI model"
- Frameworks (35 points): "React", "Next.js", "TensorFlow", "Flutter"
- Technologies (25 points): "TypeScript", "Python", "Docker", "GraphQL"

Example: "Create a React dashboard with Node.js API"
→ Score: web_application (85 points)
→ Auto-assembles: Frontend + Backend + Full Stack + QA + DevOps + MCP experts

🧠 Memory System & Session Continuity

Advanced Context Preservation:

Project-Specific Templates: Automatically creates .memory/ui-components.md for web apps, .memory/model-architecture.md for AI projects
Real-Time State Tracking: Continuous updates to project progress, decisions, and collaboration history
Auto-Compact Handling: Complete project state restoration after context compression
Session Continuity: Seamless work continuation across development sessions

🏆 Quality-First Development

Automated Quality Gates:

Quality Standards by Project Type:
web_application:
  - TypeScript coverage: 95% (auto-measured)
  - Core Web Vitals: LCP<2.5s, FID<100ms, CLS<0.1 (auto-verified)
  - Accessibility: WCAG 2.1 AA compliance (auto-checked)
  - Security: Zero vulnerabilities (auto-scanned)

ai_ml_system:
  - Model accuracy: >90% (auto-evaluated)
  - Inference latency: <100ms (auto-measured)
  - Data quality: >95% completeness (auto-verified)

🤝 Expert Collaboration System

Signal-Based Coordination:

[PROMPT_LOADED:{expert}:VERIFIED]: Expert activation with prompt validation
[COLLAB_REQUEST:{task_id}]: Cross-expert consultation requests
[TASK_DONE:{task_id}:{expert}]: Task completion with deliverable validation
Dynamic Re-engagement: Experts remain available for refinements and updates

💎 Why This Changes Everything

🎯 For Users - Simplicity at Scale

One Request = Complete System: "Create an e-commerce platform" → Full development team activated
Zero Configuration: No setup complexity, intelligent defaults handle everything
Session Continuity: Work seamlessly across days, weeks, or months
Quality Guaranteed: Enterprise-grade quality standards automatically enforced

🤖 For LLM Runtimes - Optimized Execution

Memory-Efficient: Project-specific templates minimize context usage
Execution-Focused: Concrete protocols instead of abstract documentation
Integration-Ready: All components work together seamlessly
Platform-Agnostic: Works across Claude Code, Codex, Gemini CLI, and Junie

🎮 Real-World Usage Examples

🌐 Web Application Development

User: "Create a modern e-commerce platform with React and Node.js"

AI System Response:
✅ Project Type: web_application (detected with 92% confidence)
✅ Expert Team: Frontend + Backend + Full Stack + QA + DevOps + MCP
✅ Memory Templates: ui-components.md, api-endpoints.md, user-flows.md
✅ Quality Standards: TypeScript 95%, Core Web Vitals, WCAG AA, Security scan
✅ Workflow: Requirements → Architecture → Implementation → Testing → Deployment

🤖 AI/ML System Development

User: "Build an image classification model with TensorFlow"

AI System Response:
✅ Project Type: ai_ml_system (detected with 88% confidence)
✅ Expert Team: System Dev + Research + Backend + QA + MCP
✅ Memory Templates: model-architecture.md, data-pipeline.md, training-requirements.md
✅ Quality Standards: 90% accuracy, <100ms latency, data quality validation
✅ Workflow: Research → Architecture → Model Dev → Validation → Deployment

📱 Mobile Application Development

User: "Create a React Native social media app"

AI System Response:
✅ Project Type: mobile_application (detected with 85% confidence)
✅ Expert Team: Frontend + Backend + DevOps + QA + MCP
✅ Memory Templates: platform-requirements.md, performance-targets.md
✅ Quality Standards: Platform compatibility, performance benchmarks
✅ Workflow: Requirements → Architecture → Development → Testing → Store Deploy

🔧 Advanced Configuration

🎯 Automatic System Validation

# Quick system health check
./setup-validator.sh

# Verify expert prompt loading
./validate-prompts.sh

# Test project detection
./test-detection.sh "Create a React dashboard"

🧠 Memory System Deep Dive

Intelligent Context Management:

Memory Templates by Project Type:
web_application:
  - .memory/ui-components.md      # Component library tracking
  - .memory/api-endpoints.md      # API documentation
  - .memory/user-flows.md         # UX workflow tracking

ai_ml_system:
  - .memory/model-architecture.md # Model design decisions
  - .memory/data-pipeline.md      # Data processing workflows
  - .memory/training-requirements.md # Training parameters

mobile_application:
  - .memory/platform-requirements.md # iOS/Android specifics
  - .memory/performance-targets.md   # Performance benchmarks

Auto-Compact Session Continuity:

Pre-Compact: Automatic project state preservation
Post-Restore: Complete context reconstruction with expert reactivation
Continuous Tracking: Real-time progress and decision logging

⚙️ MCP Tool Integration

Available MCP Servers:

GitHub MCP: Repository management, code examples, issue tracking
Context7: Up-to-date documentation and API references
Playwright MCP: Advanced browser automation and testing
Sequential Thinking: Complex problem-solving and analysis

Auto-Configuration:

# Claude Code setup (includes auto-compact hooks)
cp .mcp/settings-npx.json ~/.claude/settings.json
cp .claude/settings.json .claude/settings.json

# Verify MCP integration
claude-code --validate-mcp

🚀 Ready to Transform Your Development?

🌟 Get Started Today

Clone the repository:

git clone https://github.com/HanTechnology/dev-ai-agent-team-boilerplate.git
cd dev-ai-agent-team-boilerplate

Run the installer for your LLM:

./install-claude.sh    # For Claude Code (recommended)
./install-gemini.sh    # For Gemini CLI
./install-codex.sh     # For Codex
./install-junie.sh     # For Junie

Start building:

"Create a modern web application with React and Node.js"

💎 Join the AI Development Revolution

This isn't just a boilerplate—it's the future of AI-powered development. Experience:

⚡ Instant Expert Teams: Specialized AI experts assembled automatically
🧠 Intelligent Memory: Context that persists and evolves with your project
🏆 Quality Guaranteed: Enterprise-grade standards enforced automatically
🔄 Session Continuity: Seamless work across days, weeks, and months

📄 License

MIT License - Use, modify, and distribute freely. Build the future of AI development.

🙏 Acknowledgments

Powered by the convergence of advanced LLM capabilities, collaborative AI coordination, and real-world development expertise. This system represents hundreds of hours of integration testing and optimization across multiple LLM CLI runtimes.

Built for the AI development community by developers who understand the challenges of complex project coordination.

⭐ Star this repository if it transforms your development workflow!

🤝 Contribute: Help us make AI development even more powerful - PRs welcome!

💬 Community: Share your experiences and get help in our discussions

Name		Name	Last commit message	Last commit date
Latest commit History 45 Commits
.changelogs		.changelogs
.claude		.claude
.mcp		.mcp
.memory		.memory
docs		docs
prompts		prompts
workspace		workspace
.gitignore		.gitignore
.mcp.json		.mcp.json
AGENTS.md		AGENTS.md
CLAUDE.md		CLAUDE.md
GEMINI.md		GEMINI.md
README.md		README.md
backend.md		backend.md
cli-status-system.yaml		cli-status-system.yaml
deployment.md		deployment.md
export-project.bat		export-project.bat
export-project.sh		export-project.sh
frontend.md		frontend.md
fullstack.md		fullstack.md
guidelines.md		guidelines.md
install-claude.bat		install-claude.bat
install-claude.sh		install-claude.sh
install-codex.bat		install-codex.bat
install-codex.sh		install-codex.sh
install-gemini.bat		install-gemini.bat
install-gemini.sh		install-gemini.sh
install-junie.bat		install-junie.bat
install-junie.sh		install-junie.sh
install.bat		install.bat
install.sh		install.sh
logging-system.yaml		logging-system.yaml
mcp-tools.md		mcp-tools.md
memory-logging-integration.yaml		memory-logging-integration.yaml
memory-templates.yaml		memory-templates.yaml
orchestration.yaml		orchestration.yaml
project-detection.yaml		project-detection.yaml
project-management.md		project-management.md
qa-testing.md		qa-testing.md
quality-standards.json		quality-standards.json
research.md		research.md
setup-validator.sh		setup-validator.sh
system-dev.md		system-dev.md
team.yaml		team.yaml
user_req.yaml		user_req.yaml

Folders and files

Latest commit

History

Repository files navigation

Development AI Agent Team Boilerplate

🌟 What Makes This Special?

⚡ Revolutionary Features

🏗️ Core Architecture Components

🚀 LLM CLI Runtime Support

🎯 Expert Domain Specializations

📁 Complete System Architecture

🚀 Getting Started - Experience Next-Level AI Development

🎯 Simple Project Creation

📦 Installation Options

⚡ One-Command Installation

Universal Installation (Recommended)

Direct LLM-Specific Installation

🎯 How It Works - Revolutionary AI Development

🚀 Intelligent Project Detection

🧠 Memory System & Session Continuity

🏆 Quality-First Development

🤝 Expert Collaboration System

💎 Why This Changes Everything

🎯 For Users - Simplicity at Scale

🤖 For LLM Runtimes - Optimized Execution

🎮 Real-World Usage Examples

🌐 Web Application Development

🤖 AI/ML System Development

📱 Mobile Application Development

🔧 Advanced Configuration

🎯 Automatic System Validation

🧠 Memory System Deep Dive

⚙️ MCP Tool Integration

🚀 Ready to Transform Your Development?

🌟 Get Started Today

💎 Join the AI Development Revolution

📄 License

🙏 Acknowledgments

About

Resources

Uh oh!

Stars

Watchers

Forks

Releases

Packages 0

Uh oh!

Contributors

Uh oh!

Languages

Packages