1. Dynamic Model Selection
Automatically detects all your installed Ollama models Dropdown selection for easy model switching Connection status monitoring
2. AI Agents
- Code Generation Agent: Smarter code generation with better prompting
- Code Review Agent: Validates code safety and correctness
- Data Insight Agent: Provides automatic dataset analysis
3. Safety & Reliability
- Code safety validation (blocks dangerous operations)
- Error handling and user feedback
- Isolated code execution environment
4. User Experience
- Responsive UI with sidebar configuration
- Predefined example prompts for common tasks
- Real-time progress indicators
- Advanced settings panel
5. Data Processing
- Support for multiple CSV encodings
- Excel file handling
- Download options for both Excel and CSV formats
- Data preview with adjustable row count
6. Smart Code Extraction
- Code block parsing from LLM responses
- Handles various response formats
- Better code cleaning and validation
Setup Requirements:
- python 3.11++ or above
- Ollama on local system
Install required Libraries:
pip install requirments.txt
Start the ollama server inside command prompt:
ollama serve # Start Ollama server
Run the Application:
streamlit run main.py
Use the Interface:
- Select your preferred model from the dropdown
- Upload CSV/Excel files
- Use natural language prompts or example commands
- Review generated code before execution
- Download processed results
💡 Example Prompts You Can Try:
- "Add a profit margin column calculated as (revenue - cost) / revenue * 100"
- "Filter rows where sales are above the median value"
- "Create a summary table grouped by category"
- "Find all rows with missing values and highlight them"
- "Calculate running totals for the amount column"