Company Document Generator and Graph Database

This project provides tools to generate realistic company documents (contracts, invoices, payroll) and load them into a Neo4j graph database for analysis and querying.

Features

Document Generation

• Customer Contracts
• Vendor Contracts
• Customer Invoices
• Vendor Invoices
• Employee Payroll Documents
• Department Payroll Reports

Graph Database Integration

• Schema Management
• Document Ingestion
• Relationship Mapping

Prerequisites

• Devbox
• Poetry
• Neo4j (local or cloud instance)
• OpenAI API access
• Llama Parse API access

Getting Started

1. Set up Development Environment

   devbox shell
   poetry install

2. Configure Environment Variables

   Copy the example environment file and configure your settings:

   cp .env.example .env

   Required environment variables:

   • NEO4J_URI: Your Neo4j database URI (default: neo4j://localhost:7687)
   • NEO4J_USERNAME: Neo4j database username
   • NEO4J_PASSWORD: Neo4j database password
   • OPENAI_API_KEY: Your OpenAI API key for contract generation
   • LLAMA_PARSE_API_KEY: Your Llama Parse API key for document parsing

3. Generate Company Documents

   Run the following generators to create sample documents:

   # Generate customer contracts
   poetry run python -m src.generators.generate_contracts
   
   # Generate vendor contracts
   poetry run python -m src.generators.generate_vendor_contracts
   
   # Generate customer invoices
   poetry run python -m src.generators.generate_invoices
   
   # Generate vendor invoices
   poetry run python -m src.generators.generate_vendor_invoices
   
   # Generate payroll documents
   poetry run python -m src.generators.generate_payrolls

4. Set up Graph Database

   Apply the graph schema:

   poetry run python -m src.graph.apply_schema

5. Load Documents into Graph Database

   Ingest generated documents:

   poetry run python -m src.graph.loader

Generated Documents

All generated documents are stored in the company_documents directory with the following structure: