🔍 GEXA

Web Search API for AI Agents
Semantic search, web crawling, and AI-powered answers - built for the next generation of AI applications

Features • Quick Start • API • SDK • Dashboard • Docs

---

✨ Features

🔍 Semantic Search Meaning-based search powered by Google's text-embedding-004 model	🕷️ Web Crawling Playwright-based crawler with full JavaScript rendering	📄 Content Extraction Clean text & markdown extraction with Trafilatura
🤖 AI Answers Generate answers with citations using Gemini 2.5 Flash	📊 Deep Research Multi-source research reports with key points synthesis	🔗 Similar Pages Find semantically similar web pages using vector similarity

---

🚀 Quick Start

Prerequisites

Requirement	Version
Python	3.11+
PostgreSQL	14+ with pgvector
Redis	6+ (optional)

Installation

# Clone the repository
git clone https://github.com/CodeGovindz/GEXA.git
cd GEXA

# Create virtual environment
python -m venv venv
venv\Scripts\activate  # Windows
# source venv/bin/activate  # Linux/Mac

# Install dependencies
pip install -e ".[dev]"

# Install Playwright browsers
playwright install chromium

Database Setup

-- Create database and enable pgvector
CREATE DATABASE gexa;
\c gexa
CREATE EXTENSION vector;

Configuration

# Copy environment template
cp .env.example .env

# Edit .env with your configuration
# Required: GEMINI_API_KEY, DATABASE_URL

Run the Server

# Apply database migrations
alembic upgrade head

# Start the API server
$env:PYTHONPATH="src"  # Windows PowerShell
uvicorn gexa.main:app --host 127.0.0.1 --port 8000

⚠️ Windows Note: Don't use --reload flag due to Playwright subprocess compatibility

---

📡 API Endpoints

Endpoint	Method	Description
/search	POST	🔍 Semantic web search with embeddings
/contents	POST	📄 Crawl & extract content from URLs
/crawl	POST	🕷️ Background site crawling
/findsimilar	POST	🔗 Find semantically similar pages
/answer	POST	🤖 AI answers with citations
/research	POST	📊 In-depth research reports
/keys	GET/POST/DELETE	🔑 API key management

API Documentation

Once running, visit:

• Swagger UI: http://localhost:8000/docs
• ReDoc: http://localhost:8000/redoc
• Dashboard: http://localhost:8000/dashboard/

---

🐍 Python SDK

Install the SDK:

pip install gexa-py  # Coming soon to PyPI
# Or install from source:
pip install -e sdk/python

Usage

from gexa_py import Gexa

# Initialize client
gexa = Gexa("your-api-key", base_url="http://localhost:8000")

# 🔍 Semantic Search
results = gexa.search("latest AI research papers", num_results=10)

# 📄 Get Content from URLs
content = gexa.get_contents(["https://example.com"])

# 🔗 Find Similar Pages
similar = gexa.find_similar("https://openai.com")

# 🤖 Get AI Answer
answer = gexa.answer("What is transformer architecture?")

# 📊 Deep Research
research = gexa.research(
    topic="Benefits of machine learning in healthcare",
    depth="deep"
)

Async Support

from gexa_py import AsyncGexa

async def main():
    async with AsyncGexa("your-api-key") as gexa:
        results = await gexa.search("async search query")
        print(results)

---

🎨 Dashboard

GEXA includes a modern web dashboard for testing and API key management.

Access at: http://localhost:8000/dashboard/

Features

• 🧪 Playground - Test all API endpoints interactively
• 🔑 API Keys - Create, copy, and delete API keys
• 📊 Usage - Monitor API usage (coming soon)
• 📚 Docs - Quick reference and links

---

🛠️ Tech Stack

Component	Technology
Framework	FastAPI
Database	PostgreSQL + pgvector
LLM	Google Gemini 2.5 Flash
Embeddings	Google text-embedding-004
Crawler	Playwright (Chromium)
Content Extraction	Trafilatura

---

📁 Project Structure

GEXA/
├── src/gexa/           # Core API source code
│   ├── api/            # FastAPI routes
│   ├── crawler/        # Web crawling engine
│   ├── search/         # Vector search & embeddings
│   └── database/       # Models & schemas
├── sdk/python/         # Python SDK
├── dashboard/          # Web dashboard UI
├── alembic/            # Database migrations
└── tests/              # Test suite

---

🤝 Contributing

Contributions are welcome! Please feel free to submit a Pull Request.

1. Fork the repository
2. Create your feature branch (git checkout -b feature/AmazingFeature)
3. Commit your changes (git commit -m 'Add some AmazingFeature')
4. Push to the branch (git push origin feature/AmazingFeature)
5. Open a Pull Request

---

📄 License

This project is licensed under the MIT License - see the LICENSE file for details.

---

Made with ❤️ by CodeGovindz

⭐ Star this repo • 🐛 Report Bug • 💡 Request Feature