Batch Financial Compliance Engine built with Python & Streamlit
Auto-Audit Pro automates the auditing of financial documents. It ingests PDF invoices, extracts structured data using OCR, and runs them against configurable business logic (e.g., "Flag transactions > $500" or "Detect Blacklisted Vendors").
- Batch Processing: Upload 50+ invoices simultaneously for bulk auditing.
- Risk Scoring Algorithm: Automatically assigns "Low," "Medium," or "High" risk scores based on value and vendor reputation.
- Regex Pattern Matching: Detects currency values with or without symbols (e.g.,
$500or500 USD). - Interactive Dashboard: View real-time metrics (Total Value Audited, Flagged Invoices).
- Data Export: Download a structured CSV report for further analysis in Excel.
- Core: Python 3.10+
- Frontend: Streamlit
- Data Extraction: pdfplumber, RegEx
- Data Handling: Pandas
-
Clone the repository:
git clone https://github.com/Arnav184-arnav184/Auto-Audit-Pro.git
-
Navigate to the project folder:
cd Auto-Audit-Pro -
Install dependencies:
pip install -r requirements.txt
-
Run the app:
streamlit run app.py
- LLM Integration: Replace Regex with GPT-4 Vision for better spatial layout analysis on non-standard invoices.
- Database Integration: Connect to SQL to store historical audit logs rather than just CSV exports.
Created by Arnav Hazari
