Getting Started¶
Welcome to the Workflow Bank Statement Separator! This guide will help you get up and running quickly.
What You'll Learn¶
This getting started guide covers:
- Quick Start - Get running in 5 minutes
- Installation - Detailed installation instructions
- Configuration - Complete configuration guide
Prerequisites¶
Before you begin, ensure you have:
- Python 3.11+ installed on your system
- UV package manager (recommended) or pip
- OpenAI API key for optimal AI processing
- Basic command-line knowledge
UV Package Manager
We highly recommend using UV for package management. It's faster and more reliable than pip, especially for this project's complex dependencies.
System Overview¶
The Workflow Bank Statement Separator uses an 8-node LangGraph workflow to process PDF documents:
- PDF Ingestion - Load and validate documents
- Document Analysis - Extract and chunk text
- Statement Detection - AI-powered boundary identification
- Metadata Extraction - Extract account info and dates
- PDF Generation - Create separate statement files
- File Organization - Apply naming conventions
- Output Validation - Verify processing integrity
- Paperless Upload - Optional document management integration
Key Capabilities¶
Processing Features¶
- Multi-Statement PDFs: Automatically separate combined statements
- Intelligent Detection: AI-powered boundary identification
- Metadata Extraction: Account numbers, dates, bank names
- Format Preservation: Maintains original PDF formatting
Error Handling¶
- Smart Quarantine: Failed documents moved to quarantine with error reports
- Validation Levels: Configurable strictness (strict/normal/lenient)
- Recovery Suggestions: Actionable guidance for resolving issues
- Retry Logic: Automatic retry for transient failures
Integration Features¶
- Paperless-ngx: Automatic upload to document management
- Audit Logging: Complete processing trails
- CLI Management: Multi-command interface for all operations
- Configuration: 40+ environment variables for customization
Next Steps¶
Choose your path:
For immediate testing - Get running in 5 minutes
→ [Quick Start Guide](quick-start.md)
For production setup - Complete installation and configuration
→ [Installation Guide](installation.md)
For customization - Detailed configuration options
→ [Configuration Guide](configuration.md)
Support¶
Need help getting started?
- Check the Troubleshooting guide
- Review the Working Notes for detailed system information
- Report issues on GitHub