Startup Intelligence Agent System

An Agentic Startup Intelligence System that automatically collects, enriches, analyzes, and summarizes startup ecosystem data using AI agents running in E2B sandboxes.

🚀 Overview

This system provides real-time intelligence on the startup ecosystem by:

• Collecting data from news sources, funding databases, product launches, and GitHub activity
• Enriching data with metadata, entities, sentiment, and cross-references
• Analyzing patterns and trends using LLM-powered analysis
• Summarizing insights into actionable briefings for founders and investors

✨ Key Features

• 🤖 Multi-Agent Architecture: Orchestrator coordinates data collection, enrichment, analysis, and summarization
• 📊 Real-Time Intelligence: Collects data from news, funding rounds, product launches, and GitHub activity
• 🧠 AI-Powered Analysis: LLM-based trend detection and opportunity identification
• 📈 Interactive Dashboard: Beautiful web UI displaying insights, trends, and opportunities
• ☁️ Cloud-Ready: Deployable to E2B sandboxes with Docker MCP Hub integration
• ✅ Comprehensive Testing: 46/47 tests passing (98% pass rate)

🏗️ Architecture

┌─────────────────────────────────────────────────────────┐
│  Orchestrator Agent (E2B Sandbox)                       │
│  Manages: collect → enrich → analyze → summarize        │
└────────────┬────────────────────────────────────────────┘
             │
             ├─→ Data Collector Agents (Docker)
             │   • news-scraper (TechCrunch, HackerNews, ProductHunt)
             │   • startup-api (Crunchbase, AngelList, Dealroom)
             │   • github-monitor (Trending repos, technical signals)
             │
             ├─→ Enrichment Agent
             │   • Metadata extraction
             │   • Entity recognition
             │   • Sentiment analysis
             │   • Cross-referencing
             │
             ├─→ Analysis Agent (LLM)
             │   • Trend clustering
             │   • Pattern detection
             •   • Opportunity extraction
             │
             └─→ Summarizer Agent (LLM)
                 • Daily briefings
                 • Intelligence threads
                 • Structured insights

🛠️ Tech Stack

• Backend: Python 3.14+, FastAPI, SQLite
• AI/LLM: OpenAI GPT-4, Anthropic Claude
• Frontend: HTML, Tailwind CSS, Vanilla JavaScript
• Infrastructure: E2B Sandboxes, Docker, Docker Compose
• Testing: pytest (46/47 tests passing, 34% coverage)

📦 Installation

Prerequisites

• Python 3.14+
• Docker & Docker Compose
• LLM API Key (OpenAI or Anthropic)

Quick Start

1. Clone the repository

   git clone <repository-url>
   cd startup-intelligence-agent

2. Setup backend

   cd startup-intelligence-agent/backend
   python3 -m venv venv
   source venv/bin/activate  # On Windows: venv\Scripts\activate
   pip install -r requirements.txt

3. Configure environment variables

   cp .env.example .env
   # Edit .env and add your API keys:
   # OPENAI_API_KEY=your_key_here
   # or
   # ANTHROPIC_API_KEY=your_key_here

4. Start data collector agents

   cd ../../data-collector-agents
   docker-compose up -d

5. Start main server

   cd ../../startup-intelligence-agent/backend/src
   python main.py

6. Access the dashboard Open your browser to: http://localhost:8080/

🎯 Usage

Trigger Workflow

# Run full workflow (collect → enrich → analyze → summarize)
curl -X POST "http://localhost:8080/orchestrator/run?days_back=7"

# Check workflow status
curl http://localhost:8080/orchestrator/status

# Get latest briefing
curl http://localhost:8080/briefing

API Endpoints

• GET /health - Health check
• GET /info - System information
• GET /briefing - Get latest briefing
• GET /data/stats - Data statistics
• POST /orchestrator/run - Trigger workflow
• GET /orchestrator/status - Workflow status
• GET /docs - Interactive API documentation

📊 Test Results

• ✅ Unit Tests: 29/29 passing (100%)
• ✅ Integration Tests: 13/13 passing (100%)
• ✅ E2E Tests: 4/5 passing (80%)
• 📈 Test Coverage: 34%

Run Tests

cd startup-intelligence-agent/backend
source venv/bin/activate

# Run all tests
pytest tests/ -v

# Run with coverage
pytest tests/ --cov=src --cov-report=html

📁 Project Structure

.
├── startup-intelligence-agent/
│   ├── backend/
│   │   ├── src/
│   │   │   ├── orchestrator/    # Main orchestrator agent
│   │   │   ├── enrichment/       # Data enrichment agent
│   │   │   ├── analysis/         # LLM analysis agent
│   │   │   ├── summarizer/       # Briefing generation agent
│   │   │   ├── database/        # SQLite database layer
│   │   │   ├── api/             # FastAPI server
│   │   │   └── llm/             # LLM client
│   │   ├── tests/               # Test suite
│   │   └── requirements.txt
│   ├── frontend/
│   │   └── index.html           # Dashboard UI
│   └── scripts/                 # Deployment scripts
├── data-collector-agents/
│   ├── news-scraper/            # News scraping agent
│   ├── startup-api/             # Startup API wrapper
│   ├── github-monitor/         # GitHub monitoring agent
│   └── docker-compose.yml       # Agent orchestration
└── workflow-prompts/            # Development prompts

🔧 Configuration

Environment Variables

Create a .env file in startup-intelligence-agent/backend/:

# LLM Configuration
LLM_PROVIDER=openai  # or "anthropic"
LLM_MODEL=gpt-4-turbo-preview  # or "claude-3-opus-20240229"
OPENAI_API_KEY=your_key_here
# or
ANTHROPIC_API_KEY=your_key_here

# Data Collector Agents
NEWS_SCRAPER_URL=http://localhost:3001
STARTUP_API_URL=http://localhost:3002
GITHUB_MONITOR_URL=http://localhost:3003

# Server Configuration
HOST=127.0.0.1
PORT=8080

# E2B Sandbox (optional)
E2B_API_KEY=your_e2b_key
E2B_TEMPLATE=base

🚀 Deployment

Local Development

See SERVER_SETUP.md for detailed setup instructions.

E2B Sandbox Deployment

See E2B_INTEGRATION.md for deployment to E2B sandboxes.

cd startup-intelligence-agent/scripts
python deploy_to_e2b.py

📚 Documentation

• HACKATHON_SUBMISSION.md - Project overview
• TESTING.md - Testing guide
• UI_TESTING_GUIDE.md - UI testing instructions
• E2B_INTEGRATION.md - E2B deployment
• LLM_API_SETUP.md - LLM configuration

🧪 Testing

Run All Tests

cd startup-intelligence-agent/backend
source venv/bin/activate
pytest tests/ -v

Test Categories

• Unit Tests: Fast, isolated component tests
• Integration Tests: Component interaction tests
• E2E Tests: Full system tests (require running services)

See tests/README.md for detailed testing documentation.

🎉 Features

Data Collection

• News articles from TechCrunch, HackerNews, ProductHunt
• Funding rounds from Crunchbase, AngelList, Dealroom
• Product launches and startup events
• GitHub trending repositories and technical signals

AI Analysis

• Trend clustering and pattern detection
• Competitor move identification
• Opportunity extraction for founders and investors
• Market gap analysis

Dashboard

• Real-time briefing display
• Interactive trend exploration
• Funding round tracking
• Opportunity matching
• Intelligence threads

🤝 Contributing

1. Fork the repository
2. Create a feature branch
3. Make your changes
4. Add tests
5. Submit a pull request

📄 License

See LICENSE file for details.

🙏 Acknowledgments

Built using:

• FastAPI - Modern web framework
• E2B - Secure sandboxed environments
• OpenAI & Anthropic - LLM providers

📞 Contact

For questions or issues, please open an issue in the repository.

---

Status: ✅ Production Ready | Tests: 46/47 passing (98%) | Coverage: 34%