Navigate government procedures with ease using AI-powered document search and comprehensive guidance.
-
🌐 Multi-Language Support
- English, Hindi, Spanish, French, German, Urdu
- Real-time translation capabilities
-
🔍 Intelligent Search
- Advanced web crawling
- PDF document processing
- Concurrent document analysis
-
🎯 User-Friendly Interface
- Clean, modern design
- Dark mode support
- Progress tracking
- Responsive layout
-
🚀 Smart Processing
- Asynchronous operations
- Concurrent document processing
- Timeout handling
- Robust error management
# Clone the repository
git clone https://github.com/muhammadibrahim313/Llama_impact-3.2
# Navigate to project directory
cd govease
# Install dependencies
pip install -r requirements.txt
# Set up environment variables
cp .env.example .env
# Run the application
streamlit run main.py
GROQ_API_KEY=your_groq_api_key
SERPER_API_KEY=your_serper_api_key
BROWSERLESS_API_KEY=your_browserless_api_key
graph TD
A[User Query] --> B[Language Selection]
B --> C[Country Selection]
C --> D[Search Processing]
D --> E[Document Analysis]
E --> F[Translation]
F --> G[Results Display]
style A fill:#2196F3,stroke:#fff,stroke-width:2px
style G fill:#2196F3,stroke:#fff,stroke-width:2px
-
Frontend:
- Streamlit
- Custom CSS
- Responsive Design
-
Backend:
- Python 3.9+
- Asyncio
- Concurrent.futures
-
AI & Processing:
- Groq AI
- BeautifulSoup4
- PyPDF2
- Langdetect
-
APIs:
- Serper API
- Browserless API
Home Page |
Search Interface |
with concurrent.futures.ThreadPoolExecutor(max_workers=3) as executor:
future_to_url = {
executor.submit(
self.search_tool.process_url_with_timeout,
result.get('link', '')
): result.get('link', '')
for result in search_results[:total_urls]
}
gradient_css = """
.stApp {
background: linear-gradient(135deg, #1e3c72 0%, #2a5298 100%);
}
- Document Template Generation
- User Authentication System
- Mobile Application
- API Integration for Third-party Services
- Offline Document Cache
Asim Khan |
Tayyab Sajjad |
Muhammad Jawad |
Muhammad Bilal |
Muhammad Ibrahim |
Rosana Mannan |