The ultimate template for AI-assisted development with ADHD-friendly workflows and budget-conscious token usage.
A comprehensive, production-ready template that transforms how you build software with AI assistants. Designed specifically for developers who want structure, clarity, and cost control in their AI-powered development workflow.
- 🧠 ADHD-Friendly: Structured workflows that prevent overwhelm and maintain focus
- 💰 Budget-Conscious: Smart token usage tracking and cost optimization
- 🤖 AI-Optimized: Pre-configured for Claude Code, OpenAI Codex, and GitHub Copilot
- ⚡ One-Command Setup: From zero to coding in under 5 minutes
- 🎯 Task-Driven: Taskmaster AI breaks down projects into manageable chunks
- 🔄 Cross-Platform: Works on macOS, Linux, and Docker environments
# 1. Clone this template
git clone https://github.com/Cordycepsers/ai-dev-template.git
cd ai-dev-template
# 2. One-command setup (installs everything)
chmod +x /path/to/ai-dev-template/scripts/playwright_setup.sh
./scripts/python_setup.sh
./scripts/playwright_setup.sh
# 3. Add your API keys
nano .env
# 4. Create your project plan
nano .taskmaster/docs/prd.txt
# 5. Generate development tasks
task-master parse-prd .taskmaster/docs/prd.txt
# 6. Start building with AI
task-master nextThat's it! You're now ready for AI-assisted development with full task management, budget tracking, and ADHD-friendly workflows.
ai-dev-template/
├── 🎛️ .vscode/ # VS Code optimized for AI development
│ ├── settings.json # AI-friendly editor settings
│ ├── extensions.json # Essential AI development extensions
│ └── tasks.json # One-click Taskmaster integration
├── 🎯 .cursor/ # Cursor IDE configuration
│ └── mcp.json # MCP server for Taskmaster AI
├── 🤖 .claude/ # Claude Code integration
│ ├── agents/ # Specialized Claude agents
│ │ ├── task-checker.md # Quality assurance agent
│ │ ├── task-executor.md # Implementation specialist
│ │ └── task-orchestrator.md # Strategic planning agent
│ └── commands/ # Claude-specific commands
│ ├── tm # Task Master CLI wrapper
│ └── TM_COMMANDS_GUIDE.md # Command reference
├── 📋 .taskmaster/ # Task management system
│ ├── docs/ # Project documentation
│ │ └── prd.txt # Your Product Requirements Document
│ └── templates/ # PRD and task templates
│ └── prd-template.txt # Structured PRD template
├── ⚙️ config/ # Configuration management
│ ├── .env.template # Environment variables template
│ └── ai-models.json # AI model configurations & budgets
├── 📚 docs/ # Comprehensive documentation
│ ├── AGENTS.md # 🤖 AI agent personas & instructions
│ ├── WORKFLOW.md # 🧠 Complete ADHD-friendly workflow
│ ├── GITHUB_COPILOT_SETUP.md # 🔧 GitHub Copilot troubleshooting
│ ├── prompt-engineer.md # 🤖 AI prompt building helper
│ └── HOT-TO-USE-AGENTS.md
├── 🔧 scripts/ # Automation scripts
│ ├── python_setup.sh # One-command environment setup
│ └── playwright_setup.sh # One-command environment setup
├── 📊 usage-data/ # Budget tracking
│ └── monthly-usage.json # Token usage monitoring
└── 📖 README.md # This comprehensive guide
- 🍅 Pomodoro Integration: 25-minute focused work sessions
- 📝 One Task at a Time: Taskmaster ensures you never feel overwhelmed
- 🎯 Clear Next Steps: Always know exactly what to work on next
- 🏆 Progress Celebration: Visual task completion tracking
- 🤖 AI Agent Personas: Each AI has a specific role and context
- 📋 Template-Driven: Never start from a blank page
- 🔄 Consistent Workflow: Same process for every project
- 📚 Documentation-First: Everything is documented and searchable
- 📊 Task Breakdown: Complex features become simple, actionable tasks
- ⏰ Time Boxing: Built-in time limits for each activity
- 🚫 Scope Creep Protection: PRD-driven development keeps you on track
- 🎮 Gamification: Task completion provides dopamine hits
| Tool | Purpose | Configuration |
|---|---|---|
| Claude Code | Primary coding assistant | .cursor/mcp.json |
| OpenAI Codex | Code generation & review | config/ai-models.json |
| GitHub Copilot | Real-time code suggestions | .vscode/extensions.json (Setup Guide) |
| Taskmaster AI | Project management | .taskmaster/ |
| Manus | Brainstorming & documentation | Manual integration |
Each AI assistant has a specific role defined in docs/AGENTS.md:
- 🎯 PromptEngineerAgent - AI prompt creation and optimization (starting point for AI configurations)
- 🏗️ ArchitectAgent - System design and technical architecture
- 🔍 CodeReviewAgent - Code quality and best practices enforcement
- 📝 DocumentationAgent - Technical writing and documentation
- 🚀 DeploymentAgent - CI/CD and infrastructure automation
- 💰 BudgetAgent - Cost optimization and resource management
- 🧪 TestingAgent - Quality assurance and test automation
# In your AI chat:
"Please read docs/AGENTS.md for context. You are CodeReviewAgent.
Review this Python function for bugs and performance issues."- 📊 Real-time Usage Tracking - Monitor API costs as you work
- 🎯 Model Optimization - Use the right AI model for each task
⚠️ Budget Alerts - Get notified before hitting spending limits- 📈 Usage Analytics - Understand your AI spending patterns
// config/ai-models.json
{
"budget": {
"monthlyLimit": 100,
"alertThreshold": 0.8,
"usageFile": "../usage-data/monthly-usage.json"
}
}- Batch Requests: Combine multiple questions in one prompt
- Template Reuse: Avoid regenerating common patterns
- Local Tools First: Use linters/formatters before AI review
- Smart Model Selection: GPT-3.5 for simple tasks, GPT-4 for complex ones
- 🍎 macOS - Native support with Homebrew integration
- 🐧 Linux - Ubuntu and other distributions
- 🐳 Docker - Containerized development environment
- ☁️ Cloud - Works in GitHub Codespaces and similar platforms
- 🐍 Python - Conda/venv environments with testing
- 📦 Node.js - npm/yarn with modern JavaScript/TypeScript
- 🌐 Web Development - React, Vue, Flask, FastAPI templates
- ☁️ Cloud Native - Terraform, Docker, Kubernetes ready
- VS Code - Optimized settings and extensions
- Cursor IDE - AI-native development environment
- Docker - Containerized development and deployment
- Git - Version control with conventional commits
- Testing - pytest, jest, and other testing frameworks
- 🔄 WORKFLOW.md - Complete ADHD-friendly development workflow
- 🤖 AGENTS.md - AI agent personas and instructions
- 📖 HOW-TO-USE-AGENTS.md - Complete guide to using AGENTS.md effectively
- 🔧 GITHUB_COPILOT_SETUP.md - GitHub Copilot setup and troubleshooting
- 🛠️ VSCODE_TROUBLESHOOTING.md - Fix VS Code reloading, extension errors, and performance issues
- 🎯 .claude/ - Claude Code integration with specialized agents and commands
- Setup Scripts - Platform-specific automated setup
- 💡 Ideation - Brainstorm with AI and research tools
- 📋 Planning - Create structured PRD with AI assistance
- 🔧 Setup - One-command environment configuration
- 🎯 Development - Task-driven coding with AI assistants
- 🧪 Testing - Automated testing and quality assurance
- 🚀 Deployment - Containerized deployment and CI/CD
- Solo Developers building side projects
- Startup Teams needing rapid prototyping
- Learning Developers who want AI guidance
- ADHD Developers who need structure and focus
- Budget-Conscious Teams tracking AI costs
- Web Applications (React, Vue, Flask, FastAPI)
- APIs and Microservices (REST, GraphQL)
- Data Science Projects (Jupyter, pandas, scikit-learn)
- Cloud Applications (AWS, GCP, Azure)
- Mobile Apps (React Native, Flutter)
# Parse your PRD into actionable tasks
task-master parse-prd .taskmaster/docs/prd.txt
# See what to work on next
task-master next
# Break down complex tasks
task-master expand --id=5 --num=3
# Track your progress
task-master list --status=done# Build and test locally
docker build -t my-project .
docker run -p 8000:8000 my-project
# Full development environment
docker-compose up -d- Ctrl+Shift+P → "Tasks: Run Task" → "Taskmaster: Next Task"
- Ctrl+Shift+P → "Tasks: Run Task" → "Python: Run Tests"
- Ctrl+Shift+P → "Tasks: Run Task" → "Docker: Build"
This template is designed to be your foundation. Customize it by:
- Adding new AI agent personas for your specific domain
- Extending the setup script for your preferred tools
- Creating project-specific templates for common patterns
- Sharing improvements back to the community
MIT License - Use this template for any project, commercial or personal.
# Start your AI-powered development journey
git clone <this-repo> my-awesome-project
cd my-awesome-project
./scripts/setup.shWelcome to the future of development! 🚀🤖