Transform your ideas into reality 10x faster—while keeping your unique creative vision.
This isn't just another project template. It's a complete AI-first development environment that amplifies your capabilities while preserving what makes your work uniquely yours. It is currently configured to work with Claude Code and Gemini CLI.
- 🔔 Never break flow with desktop notifications when AI needs input
- ✅ Stop thinking about quality with automated checks on every change
- 🧠 Code your way with philosophy-driven development built-in
- 📚 Feed AI perfectly with smart context management tools
- ⚡ Start building in 5 minutes instead of setting up for hours
You can. Starting right now.
# 1. Create from template
Click the "Use this template" button on GitHub to create a new repository.
# 2. Clone your new repository
git clone https://github.com/[your-username]/[your-new-repo-name]
cd [your-new-repo-name]
# 3. In your AI assistant, run:
/prime
# 4. Start building!
/ultrathink-task Help me build a web app that [your amazing idea here]That's it. You're now developing at AI velocity with human creativity.
Want to dive deeper? Check out this auto-generated interactive documentation site!
Finally, you can prototype as fast as you can imagine. Natural language to working code in minutes.
Stop context switching. Automated quality checks and smart notifications keep you in deep work.
Build platforms, not just projects. Your expertise becomes your team's capability.
Move at the speed of thought. Test ideas in hours, not weeks.
Never check your terminal again. Get native notifications when:
- Claude Code needs permission to proceed
- Tasks complete successfully
- Errors need your attention
Works seamlessly on Mac, Linux, Windows, and WSL.
The /ultrathink-task command orchestrates specialized AI agents:
- Architect Agent: Designs high-level approach
- Research Agent: Gathers knowledge and best practices
- Coder Agent: Implements solutions
- Tester Agent: Validates and verifies
Complex problems solved systematically, every time.
Every code change triggers quality checks automatically:
- Formatting fixed
- Linting applied
- Type checking run
- Tests executed
You focus on creating. Quality happens automatically.
Your development principles embedded in every interaction:
- Simplicity First: Clean, maintainable code by default
- Human-Centric: AI amplifies, never replaces your creativity
- Pragmatic Choices: Real-world solutions, not academic exercises
Feed AI the right information at the right time:
ai_context/- Store persistent reference docsai_working/- Workspace for AI planning and iteration- Auto-generate project documentation
- Fetch external library docs on demand
- Python/TypeScript pre-configured with modern tools
- Recursive Make system for monorepos
- Git-ignored temp spaces for experimentation
- Cross-platform compatibility guaranteed
- ✅
/ultrathink-task- Multi-agent orchestration for complex problems - ✅
/prime- Philosophy-aligned environment setup - ✅
/test-webapp-ui- Automated UI testing with browser control - ✅ MCP Servers - Extended capabilities (docs lookup, browser automation)
- ✅ Custom command framework - Build your own workflows
- ✅ Desktop notifications - Stay in flow (all platforms)
- ✅ Auto quality checks - Format, lint, type-check automatically
- ✅ Smart Make system - Works with monorepos
- ✅ Pre-configured permissions - Security with productivity
- ✅ Philosophy documents - Guide AI to code your way
- ✅ AI context tools - Generate and manage documentation
- ✅ Git collector - Fetch external library docs
- ✅ Working spaces - Persistent and temporary AI workspaces
- ✅ File collectors - Smart project documentation generation
- An AI assistant with CLI capabilities:
- Python 3.11+ or Node.js 18+
- Git
- Click the Use this template button on the GitHub repository page.
- Give your new repository a name and create it.
- Clone your new repository to your local machine:
git clone https://github.com/[your-username]/[your-new-repo-name].git
cd [your-new-repo-name]# In your AI assistant, use the command:
/prime
# This will:
# - Install all dependencies
# - Activate your environment
# - Load philosophy documents
# - Run initial checks
# - Confirm everything works
/ultrathink-task Create a simple web API that:
- Has a health check endpoint
- Stores data in memory
- Returns JSON responses
- Includes proper error handling
- Has tests
Follow our philosophy of simplicity and pragmatism.
Watch as multiple AI agents collaborate to build your API with tests, error handling, and documentation.
Make any code change and watch:
- 🔔 Desktop notification when complete
- ✅ Automatic quality checks run
- 📝 Clear feedback in your terminal
- 🎯 Stay focused on your next idea
Your New Superpower: Prototype at the Speed of Thought
/ultrathink-task I need a dashboard that shows:
- User growth over time with a nice chart
- Current active users in real-time
- Revenue metrics with month-over-month change
- Mobile responsive design
- Export to PDF functionality
Make it look professional but keep it simple.
What Happens Next:
- Architect designs the component structure
- Research finds best charting libraries
- Coder implements with your requirements
- Tester ensures it all works
- You get a working dashboard in minutes
Pro Tips:
- Drop mockups in
ai_working/for reference - Use
/test-webapp-uito see it in action - Iterate with natural language: "Make the charts bigger"
Your New Reality: Deep Work Without Interruptions
# Set up context for your feature
echo "API Schema: ..." > ai_context/api-spec.md
echo "Database Models: ..." > ai_context/models.md
# Let AI help with implementation
/ultrathink-task Implement the user authentication system based on:
@ai_context/api-spec.md
@ai_context/models.md
Use JWT tokens, include refresh logic, and add rate limiting.Automation That Backs You Up:
- Every save triggers quality checks
- Desktop notifications for important events
- Philosophy guides ensure consistent code style
- Context persistence across sessions
Level Up Your Workflow:
- Create custom commands for repetitive tasks
- Add project-specific hooks for your stack
- Share configs with your team via Git
Your New Platform: Scalable AI-Assisted Development
# Create team-wide commands
# For Claude Code:
cat > .claude/commands/create-microservice.md << 'EOF'
## Usage
`/create-microservice <service-name>`
## Process
1. Create service directory structure
2. Set up Docker configuration
3. Create base API with health checks
4. Set up testing framework
5. Add to docker-compose
6. Create README with setup instructions
EOF
# For Gemini CLI:
cat > .gemini/commands/create-microservice.toml << 'EOF'
description = "Create a new microservice"
prompt = """
## Usage
`/create-microservice <service-name>`
## Process
1. Create service directory structure
2. Set up Docker configuration
3. Create base API with health checks
4. Set up testing framework
5. Add to docker-compose
6. Create README with setup instructions
"""
EOF
# Share with your team
git add .claude/ .gemini/
git commit -m "Add team microservice generator"
git pushBuild Your Platform:
- Encode architectural decisions in commands
- Automate compliance and standards
- Share expertise through configuration
- Scale your impact across teams
This template embodies a philosophy of "Human Creativity, AI Velocity":
-
You Are The Visionary
- AI handles implementation details
- You focus on what to build, not how
- Your creativity remains uniquely yours
-
Philosophy Over Process
- Principles guide every decision
- Simplicity beats complexity
- Pragmatism over perfection
-
Flow State Is Sacred
- No unnecessary interruptions
- Automation handles the mundane
- Notifications only when essential
-
Built To Share
- Your setup helps your team
- Knowledge embedded in tools
- Success patterns spread naturally
Read more in Philosophy Deep Dive
# 1. Set up context
mkdir -p ai_working/feature-x
echo "Feature requirements..." > ai_working/feature-x/spec.md
# 2. Prime your environment
/prime
# 3. Build with guidance
/ultrathink-task Implement feature X based on @ai_working/feature-x/spec.md# 1. Capture the problem
echo "Error details..." > ai_working/tmp/debug-notes.md
# 2. Analyze systematically
/ultrathink-task Debug this issue:
- Error: [paste error]
- Context: @ai_working/tmp/debug-notes.md
- Check our patterns in @ai_context/IMPLEMENTATION_PHILOSOPHY.md# 1. Describe what you want
/ultrathink-task Create a data table component that:
- Handles large datasets efficiently
- Has sorting and filtering
- Exports to CSV
- Looks good on mobile
# 2. Test it immediately
/test-webapp-ui
# 3. Iterate quickly
Actually, add pagination and make the rows clickable- Edit
ai_context/IMPLEMENTATION_PHILOSOPHY.md - Add domain-specific principles
- Include coding standards
- Document decision patterns
- Create
.claude/commands/your-command.mdfor Claude Code - Create
.gemini/commands/your-command.tomlfor Gemini
See existing commands for examples
- Edit
.claude/tools/notify.shfor custom messages - Adjust notification triggers in settings
- Add sound alerts if desired
In .claude/settings.json:
{
"hooks": {
"PostToolUse": [
{
"matcher": "YourCustomTool",
"hooks": [{ "command": "your-script.sh" }]
}
]
}
}- Command Reference - All commands explained
- Automation Guide - Hooks and quality checks
- Context Management - Feeding AI effectively
- Philosophy Guide - Why this approach works
You're 5 minutes away from developing at a new level. Your creativity, amplified by AI, guided by philosophy, and backed by automation.
The future of development isn't about AI replacing developers. It's about developers achieving what was previously impossible.
Welcome to AI-amplified development. Let's build something amazing.
Made with ❤️ by developers, for developers
Star this repo if it helps you build faster