⚡ We Verify

AI-Powered Startup Idea Validation Engine

Validate any startup idea in under 2 minutes with parallel AI agents, real-time web research, and a consultant-style scored report.

---

🚀 What It Does

Submit a startup idea → 3 AI agents run in parallel → get a consultant-style validation report with scores, evidence, competitor analysis, and actionable next steps.

Agent	Role
📊 Market Agent	Searches the web for market demand signals, TAM estimates, and growth trends
🏢 Competitor Agent	Identifies real competitors, their pricing, features, strengths, and weaknesses
🧠 Synthesizer	Scores the idea across 4 dimensions and delivers a final verdict

Verdicts

• ✅ VALIDATE — Strong signals, worth pursuing
• 🔄 REWORK — Promising but needs refinement
• 🛑 DO NOT BUILD — Weak signal, reconsider or pivot

---

🛠️ Tech Stack

Layer	Technology
Frontend	React 18 + Vite, NDJSON stream reader
Backend	Express.js, parallel agent orchestration
LLM	OpenRouter API (LLaMA 3 70B)
Search	Tavily API (real-time web search)
Database	Supabase (PostgreSQL)
Streaming	Chunked NDJSON over HTTP POST

---

⚡ Quick Start

# 1. Clone
git clone https://github.com/Suryabarsaiya/we-verify.git
cd we-verify

# 2. Set up environment
cp .env.example .env
# Fill in your API keys (OpenRouter, Tavily, Supabase)

# 3. Install dependencies
npm run install-all

# 4. Run
npm run dev

Open http://localhost:5173 in your browser.

---

🔑 Environment Variables

Variable	Required	Description
OPENROUTER_API_KEY	✅	OpenRouter API key for LLM access
OPENROUTER_MODEL	✅	Model ID (default: meta-llama/llama-3-70b-instruct:nitro)
TAVILY_API_KEY	✅	Tavily API key for web search
SUPABASE_URL	Optional	Supabase project URL for logging
SUPABASE_KEY	Optional	Supabase anon/public key
PORT	Optional	Server port (default: 3001)

See .env.example for the template.

---

📁 Project Structure

we-verify/
├── server/
│   ├── agents/
│   │   ├── orchestrator.js      # Parallel agent dispatch + pipeline trace
│   │   ├── marketAgent.js       # Market demand & trend analysis
│   │   ├── competitorAgent.js   # Competitor identification & mapping
│   │   └── synthesizerAgent.js  # Score aggregation & verdict
│   ├── services/
│   │   ├── llm.js               # OpenRouter LLM (callLLM, parseJSON, prompts)
│   │   ├── search.js            # Tavily web search
│   │   └── supabase.js          # Database logging
│   ├── routes/
│   │   └── api.js               # POST /api/validate (NDJSON stream)
│   ├── tests/                   # Jest + Supertest
│   └── index.js                 # Express entry point
├── client/
│   └── src/
│       ├── components/
│       │   ├── IdeaForm.jsx         # Input form
│       │   ├── AgentProgress.jsx    # Live terminal stream
│       │   ├── ValidationReport.jsx # Bento-box report
│       │   ├── ScoreGauge.jsx       # Circular score gauge
│       │   └── CompetitorCard.jsx   # Competitor display card
│       ├── App.jsx              # NDJSON stream handler + routing
│       └── App.css              # Futuristic SaaS theme
├── .env.example
├── .gitignore
├── LICENSE
├── CONTRIBUTING.md
└── package.json

---

🧪 Testing

# Backend tests (Jest + Supertest)
npm run test:backend

# Frontend tests (Vitest)
npm run test:frontend

---

🤝 Contributing

See CONTRIBUTING.md for setup instructions and guidelines.

---

📄 License

MIT — built by @Suryabarsaiya