Owlbase.ai MVP
PietroPavesi/owlbase-mvp
Converta expertise de especialistas em documentação e treinamento via IA
Ver no GitHubPlataforma de captura de conhecimento tácito que transforma conversas com especialistas em documentação profissional, materiais de treinamento e ativos de conhecimento prontos para negócios. Integrada com Claude AI, automatiza a geração de documentos, árvores de decisão e datasets estruturados através de entrevistas inteligentes e adaptativas.
Instalação
npx convez-ai PietroPavesi/owlbase-mvp1 comando no terminal: detecta o formato do repositório (marketplace ou plugin avulso) e instala via Claude Code. Requer Node 18+ e o Claude Code instalado.
Métricas
Casos de uso
- Capturar conhecimento crítico antes de aposentadorias ou saídas de funcionários
- Gerar documentação de processos e manuais de treinamento automaticamente
- Estruturar dados para treinar modelos de IA customizados
- Criar bases de conhecimento para PMEs sem expertise em documentação
Veredito

MVP sólido com fluxo de descoberta e integração Claude funcionando. Diferencial real está na automatização de documentação a partir de conversas estruturadas — resolver bem a fase de geração de documentos será crítico para o produto decolar.
README
A tradução para o português deste README ainda está na fila.
Updated README.md:
# Owlbase.ai MVP
> Transform expert knowledge into professional documentation through AI-powered conversations
**STATUS: ✅ Core MVP Complete - Chat interface with Claude AI integration working**
Owlbase.ai helps SME leaders capture critical knowledge before employees retire or leave. Through natural 30-minute conversations with Owly (our AI assistant), experts automatically generate process documents, training materials, decision trees, and AI-ready datasets.
## 🎯 Problem We're Solving
- **42%** of domain knowledge exists only in employees' minds
- **10,000** Baby Boomers retire daily, taking decades of expertise with them
- **75%** of companies struggle to implement AI due to lack of structured knowledge
## 🦉 How Owlbase Works
1. **Discovery Phase** - Owly asks 4 structured questions to understand the expertise area
2. **AI Interview** - Claude AI conducts an intelligent, adaptive interview based on discovery context
3. **Document Generation** - AI transforms conversation into professional business documentation *(coming next)*
## 🚀 Current Features (MVP v0.2.0)
### ✅ Completed
- **Admin Panel** - Create clients and authorize expert email addresses
- **Expert Authentication** - Secure email-based access control
- **Profile Setup** - Comprehensive expert background capture
- **Discovery Interview** - 4 structured questions to gather process context
- **AI-Powered Interview** - Claude AI integration for intelligent follow-up questions
- **Two-Phase Flow** - Smooth transition from discovery to AI interview
- **Session Management** - Real-time chat with conversation persistence
- **Branded Interface** - Professional UI following Owlbase design system
### 🔄 In Development
- **Document Generation** - Auto-create process docs, training materials, decision trees
- **Knowledge Assets Library** - View, download, and manage generated documents
- **Session History** - Review and access previous interviews
- **Document Templates** - Customizable output formats
### 📋 Planned Features
- **Multi-session Projects** - Break large knowledge areas into multiple sessions
- **Collaborative Review** - Team feedback on generated documents
- **Export Options** - PDF, Word, Confluence integration
- **Analytics Dashboard** - Track knowledge capture progress
- **AI Training Data Export** - Structured datasets for custom AI models
## 🛠 Tech Stack
- **Frontend**: Next.js 15 (App Router), TypeScript, Tailwind CSS v4
- **Backend**: Supabase (PostgreSQL + Auth + RLS)
- **AI**: Anthropic Claude 3.5 Sonnet for conversations
- **Deployment**: Vercel (ready)
## 📋 Database Schema
```sql
clients - SME customer companies
├── allowed_experts - Email whitelist per client
├── experts - Registered expert profiles
├── sessions - Interview conversations with Owly
└── documents - Generated knowledge assets (coming next)
🏗 Development Setup
# Clone repository
git clone https://github.com/yourusername/owlbase-mvp.git
cd owlbase-mvp
# Install dependencies
npm install
# Set up environment variables
cp .env.example .env.local
# Add your Supabase and Anthropic API keys
# Run development server
npm run dev
Environment Variables
NEXT_PUBLIC_SUPABASE_URL=your_supabase_project_url
NEXT_PUBLIC_SUPABASE_ANON_KEY=your_supabase_anon_key
ANTHROPIC_API_KEY=your_anthropic_api_key
🎨 Brand Guidelines
Owlbase follows a warm, professional design system:
- Primary Colors: Amber Glow (#FF8C00), Sage Wisdom (#059669)
- Typography: Inter (headings), Source Serif Pro (wisdom moments)
- Voice: Wise mentor and trusted colleague
- Logo: Owl icon representing wisdom and guidance
📱 User Flows
Admin Flow ✅
- Visit
/admin - Create new client company
- Add authorized expert email addresses
- Share platform access with experts
Expert Flow ✅
- Visit
/dashboard - Enter authorized email address
- Complete profile setup (first time)
- Click "Start New Session with Owly"
- Discovery Phase: Answer 4 structured questions about expertise area
- AI Interview: Engage in intelligent conversation with Claude AI
- View generated documents (coming next)
Interview Experience ✅
The two-phase interview system ensures comprehensive knowledge capture:
Phase 1: Discovery (4 Questions)
- What process/expertise to focus on?
- Who's involved, frequency, triggers, goals?
- Main steps, tools, challenges, skills?
- Decisions, mistakes, unwritten rules, success measures?
Phase 2: AI Interview
- Claude AI takes over with intelligent follow-up questions
- Adapts based on discovery context and expert profile
- Explores edge cases, decision frameworks, practical wisdom
- Uncovers tacit knowledge and "war stories"
🔐 Security
- Row Level Security (RLS) policies for data isolation
- Email-based access control (no passwords for MVP)
- Client-specific expert authorization
- Secure API key management
- Conversation data encrypted at rest
🚀 Deployment
# Build for production
npm run build
# Deploy to Vercel
vercel deploy
Set environment variables in Vercel dashboard.
📊 Business Model
- Target: SME (Small/Medium Enterprises)
- Pricing: $2,000-3,500 per project
- Model: Project-based consulting → SaaS transition planned
- Timeline: 6 months to launch, then find co-founder
🧪 Testing the MVP
- Setup: Create client in
/admin, add expert emails - Expert Flow: Login at
/dashboard, complete profile - Interview: Start session, go through discovery + AI interview
- Conversation: Test Claude's knowledge extraction capabilities
🎯 Next Milestones
v0.3.0 - Document Generation (Next Sprint)
- Document generation API
- Process documentation templates
- Training material creation
- Decision tree generation
- Download/export functionality
v0.4.0 - Knowledge Management
- Session history dashboard
- Document library interface
- Search and filter capabilities
- Document versioning
v1.0.0 - Production Ready
- Performance optimization
- Advanced security features
- Multi-client dashboard
- Payment integration
- Customer onboarding automation
🤝 Contributing
This is an MVP in active development. Current focus areas:
- Document generation pipeline
- User experience optimization
- Performance improvements
- Error handling and edge cases
📈 Recent Updates
v0.2.0 (Current) - Chat Interface Complete
- Two-phase interview system working
- Claude AI integration functional
- Real-time conversation with persistence
- Expert profile context integration
v0.1.0 - Foundation
- Database schema and authentication
- Branded admin and expert interfaces
- Profile management system
📄 License
Proprietary - All rights reserved
Built with ❤️ for the future of knowledge work
Owlbase.ai - Where expertise becomes organizational wisdom
🦉 Ready for beta testing with real SME clients