Transform your GitHub repositories into an AI-powered project management hub. Solo-dev organization, finally figured out.
Stop hunting down project timelines. Stop scrambling for client meeting summaries. This system reads your actual repositories and gives you:
- 🧠 AI Repository Analysis: Deep understanding of your codebase structure, tech stack, and code quality
- 📝 Smart Todo Generation: AI creates actionable task lists based on your actual code, not generic templates
- 📊 Automated Project Recaps: Perfect meeting summaries with real GitHub metrics
- 💬 Repository Chat: Ask your codebase questions and get intelligent answers
- 🔗 Full REST API: Integrate with Cursor IDE, Slack, or anywhere else you need project data
- 🎨 Beautiful UI: Glassmorphic dark theme that doesn't hurt your eyes during late-night coding
git clone https://github.com/mbishopfx/bishoptech-githubhelper.git
cd bishoptech-githubhelper
npm installYou'll need these services (all have free tiers):
- Go to OpenAI Platform
- Create a new API key
- Copy it (starts with
sk-)
- Create account at Supabase
- Create a new project
- Go to Settings → API
- Copy your Project URL and anon public key
- Go to Settings → API → Service Role keys
- Copy your service_role key (keep this secret!)
- Go to GitHub Settings → Personal Access Tokens
- Create new token (classic)
- Select scopes:
repo,read:user,read:org - Copy your token (starts with
ghp_orgithub_pat_)
# Copy the environment template
cp .env.local.example .env.localEdit .env.local with your keys:
# OpenAI Configuration
OPENAI_API_KEY=sk-your-openai-key-here
# Supabase Configuration
NEXT_PUBLIC_SUPABASE_URL=https://your-project.supabase.co
NEXT_PUBLIC_SUPABASE_ANON_KEY=your-anon-key-here
SUPABASE_SERVICE_ROLE_KEY=your-service-role-key-here
# GitHub Integration
GITHUB_TOKEN=your-github-token-here
# App Security (generate a random 32+ character string)
NEXTAUTH_SECRET=your-random-secret-string-here
# Single User Configuration (personalize it!)
SINGLE_USER_EMAIL=[email protected]
SINGLE_USER_GITHUB_USERNAME=your-github-username
SINGLE_USER_NAME=Your Display Name
# API Security (for integrations)
MASTER_API_KEY=gha_your-generated-api-key# Copy the database setup script
# Go to your Supabase project → SQL Editor
# Paste and run the contents of: database/quick-setup.sqlOr run it manually:
- Open Supabase SQL Editor
- Copy contents from
database/quick-setup.sql - Run the script
- Verify tables were created in Table Editor
npm run devOpen http://localhost:3000 - you should see the GitHub Helper dashboard!
- Next.js 15 API Routes: Rock-solid
/api/v1/endpoints - LangGraph Multi-Agent System: AI agents that specialize in different tasks
- GPT-4o Integration: Context-aware AI that actually understands your code
- Supabase PostgreSQL: Scalable database with Row Level Security
- GitHub API via Octokit: Real-time repository data and analysis
- Single-User Auth: Simplified security model for personal/small team use
- Glassmorphic UI: macOS-inspired dark theme that looks professional
- Framer Motion: Smooth animations and micro-interactions
- TypeScript: Full type safety across the entire application
- Tailwind CSS: Utility-first styling with custom glassmorphic components
- Real-time Streaming: AI responses stream in real-time, no loading screens
- Mobile Responsive: Works great on phones and tablets
src/
├── app/
│ ├── api/v1/ # Versioned API endpoints
│ │ ├── ai/chat/ # AI chat functionality
│ │ ├── projects/ # Repository management
│ │ ├── todos/ # Task management
│ │ └── recaps/ # Project summaries
│ └── dashboard/ # UI pages
│ ├── chat/ # Chat interface
│ ├── todos/ # Task management
│ └── repositories/ # Repository browser
├── lib/
│ ├── agents/ # AI agent implementations
│ │ ├── graphs/ # LangGraph workflow definitions
│ │ └── tools.ts # AI tools and functions
│ ├── github.ts # GitHub API integration
│ └── supabase.ts # Database client
└── components/ui/ # Reusable UI components
- Navigate to Repositories in the dashboard
- Click Import from GitHub
- Your repositories will be analyzed automatically
- Review the AI-generated insights
- Select any repository
- Click Generate Todo List
- AI analyzes your code and creates actionable tasks
- Tasks are prioritized based on code complexity and dependencies
- Go to Recaps section
- Select date range and repositories
- AI generates meeting-ready summaries with:
- Progress metrics from GitHub
- Recent commits and changes
- Identified issues and recommendations
- Next steps and priorities
- Open any repository in the Chat interface
- Ask questions like:
- "What's the main architecture of this project?"
- "What dependencies need updating?"
- "What would be the best way to add user authentication?"
- AI provides context-aware answers based on your actual code
Add this to your Cursor rules:
# GitHub Helper API
LLM Integration: https://your-deployment.vercel.app/llm.txt
// Generate AI todos for any project
const response = await fetch('/api/v1/todos', {
method: 'POST',
headers: {
'Authorization': 'Bearer gha_your_api_key',
'Content-Type': 'application/json'
},
body: JSON.stringify({
title: 'Sprint Planning Tasks',
ai_generate: true,
context_prompt: 'Focus on testing and performance improvements'
})
});
// Get project recap
const recap = await fetch('/api/v1/recaps', {
method: 'POST',
headers: { 'Authorization': 'Bearer gha_your_api_key' },
body: JSON.stringify({
period: 'weekly',
include_metrics: true
})
});# Install Vercel CLI
npm i -g vercel
# Deploy
vercel --prodAdd these in your Vercel dashboard:
NODE_ENV=production
OPENAI_API_KEY=your_production_openai_key
NEXT_PUBLIC_SUPABASE_URL=your_production_supabase_url
NEXT_PUBLIC_SUPABASE_ANON_KEY=your_production_supabase_anon_key
SUPABASE_SERVICE_ROLE_KEY=your_production_service_key
GITHUB_TOKEN=your_production_github_token
NEXTAUTH_SECRET=your_production_secret_minimum_32_chars
[email protected]
SINGLE_USER_GITHUB_USERNAME=your-github-username
SINGLE_USER_NAME=Your Display Name
MASTER_API_KEY=gha_your_production_api_key
Edit src/lib/agents/config.ts:
export const AI_CONFIG = {
model: 'gpt-4o', // Change AI model
temperature: 0.7, // Adjust creativity (0.0-1.0)
maxTokens: 2000, // Response length limit
};Modify tailwind.config.ts and src/app/globals.css for your branding.
Create new graphs in src/lib/agents/graphs/ using the LangGraph framework.
"Can't connect to Supabase"
- Verify your
NEXT_PUBLIC_SUPABASE_URLandNEXT_PUBLIC_SUPABASE_ANON_KEY - Make sure you ran the database setup script
- Check that your Supabase project is active (not paused)
"GitHub API rate limit exceeded"
- Your
GITHUB_TOKENmight be invalid or expired - Create a new token with proper scopes:
repo,read:user,read:org - GitHub has rate limits - wait an hour or upgrade to GitHub Pro
"OpenAI API errors"
- Check your
OPENAI_API_KEYis correct (starts withsk-) - Verify you have credits in your OpenAI account
- Some features require GPT-4 access
"Database errors"
- Run
database/quick-setup.sqlagain in Supabase SQL Editor - Check that all tables were created in your Supabase Table Editor
- Verify your
SUPABASE_SERVICE_ROLE_KEYhas full database access
If you're stuck:
- Check the Issues tab for similar problems
- Create a new issue with:
- Your operating system
- Node.js version (
node --version) - Error message (remove any API keys!)
- Steps you've tried
MIT License - feel free to fork, modify, and use in your own projects!
- Next.js 15 - React framework
- OpenAI GPT-4o - AI language model
- LangGraph - Multi-agent AI workflows
- Supabase - PostgreSQL database and auth
- Tailwind CSS - Utility-first CSS framework
- Framer Motion - Animation library
- GitHub API - Repository data access
⭐ Star this repo if it helps you get organized!
🐛 Found a bug? Report it here