Audit-Canvas

AI-Assisted Audit Data Quality & Reconciliation Platform

This is the Audit Canvas Project for IBM Bob Dev Hackathon - a web-based AI-assisted audit platform designed to streamline data validation, reconciliation, and audit documentation.

🚀 Quick Start

Prerequisites

Python 3.11.3 or higher
pip (Python package manager)
Windows OS (for .bat scripts) or Linux/Mac

Installation

Clone the repository

git clone <repository-url>
cd Audit-Canvas

Install all required packages
```
pip install -r requirements.txt
```
Run security vulnerability scan
```
pip-audit --desc
```

Fix security vulnerabilities (IMPORTANT!)

# Windows
fix_vulnerabilities.bat

# Linux/Mac
chmod +x fix_vulnerabilities.sh
./fix_vulnerabilities.sh

📦 Installed Libraries

Core Framework & Data Processing

streamlit (1.57.0) - Web UI framework
pandas (3.0.2) - Data manipulation and analysis
numpy (2.4.4) - Numerical computing

File Handling & Excel Support

openpyxl (3.1.5) - Excel (.xlsx) file operations
xlrd (2.0.2) - Excel (.xls) file reading
xlsxwriter (3.2.9) - Enhanced Excel report generation

Data Quality & Analysis

python-dateutil (2.9.0.post0) - Date validation
scipy (1.17.1) - Statistical analysis
plotly (6.7.0) - Interactive data visualization

Report Generation

jinja2 (3.1.6) - HTML template rendering

Security & Monitoring

pip-audit (2.10.0) - Vulnerability scanner

Optional Enhancements

streamlit-aggrid (1.2.1.post2) - Enhanced data grids
pillow (12.2.0) - Image handling
requests (2.32.5) - HTTP operations

For a complete list, see LIBRARIES_SUMMARY.md

🔒 Security

Vulnerability Status

✅ SECURE: All vulnerabilities have been resolved! 0 known vulnerabilities

Security Status (Updated 2026-05-02 20:43 UTC):

✅ All 18 vulnerabilities have been patched
✅ All packages upgraded to secure versions
✅ Verified with pip-audit: No known vulnerabilities found

Remediation Summary:

🔴 Critical: 1 vulnerability → ✅ FIXED (mysql-connector-python 9.1.0)
🟠 High: 10 vulnerabilities → ✅ FIXED (urllib3 2.6.3, setuptools 78.1.1, protobuf 6.33.5)
🟡 Medium: 5 vulnerabilities → ✅ FIXED (pip 26.1, python-dotenv 1.2.2, requests 2.33.0)
🟢 Low: 2 vulnerabilities → ✅ FIXED (pygments 2.20.0, pytest 9.0.3)

Security Documentation

All security files are organized in the security-docs/ folder:

security-docs/README.md - START HERE - Complete guide to all security files
security-docs/SECURITY_REMEDIATION_LOG.md - Complete remediation log
security-docs/SECURITY_VULNERABILITY_REPORT.md - Initial findings
security-docs/vulnerability_scan.json - Machine-readable scan results
security-docs/fix_vulnerabilities.bat - Automated fix script

Regular Security Scanning

# Run weekly vulnerability scans
pip-audit --desc

# Generate JSON report
pip-audit --format json --output vulnerability_scan.json

🎯 Platform Features

Based on the AI Audit Platform Design Document:

1. Data Profiling

Row/column counts
Missing values detection
Duplicate detection
Data type validation
Outlier identification

2. Data Quality Rule Engine

Required fields validation
Unique primary keys
Date format validation
Value range checks
Referential integrity

3. Reconciliation Engine

Record count comparison
Missing/extra records detection
Column mapping validation
Data type consistency checks

4. AI-Powered Features

Anomaly explanation in plain English
Column mapping suggestions
Automated audit findings generation
Documentation assistance

5. Report Generation

Excel workpapers export
HTML audit reports
Interactive visualizations

🏗️ Technology Stack

Frontend: Streamlit
Backend: Python 3.11.3
Data Processing: pandas, numpy, scipy
Visualization: plotly
AI Integration: IBM Bob
Output Formats: Excel (openpyxl, xlsxwriter), HTML (jinja2)

📋 Project Structure

Audit-Canvas/
├── AI_Audit_Platform_Design_Document_Full.pdf  # Full design specification
├── requirements.txt                             # Python dependencies (updated with secure versions)
├── LIBRARIES_SUMMARY.md                         # Detailed library documentation
├── LICENSE                                      # Project license
├── README.md                                    # This file (main documentation)
│
└── security-docs/                               # Security documentation folder
    ├── README.md                                # Guide to all security files
    ├── SECURITY_VULNERABILITY_REPORT.md         # Initial vulnerability analysis
    ├── SECURITY_REMEDIATION_LOG.md              # Complete remediation timeline
    ├── vulnerability_scan.json                  # Machine-readable scan results
    └── fix_vulnerabilities.bat                  # Automated remediation script

🚦 Getting Started with Development

1. Set Up Environment

# Verify Python version
python --version  # Should be 3.11.3 or higher

# Install dependencies
pip install -r requirements.txt

# Fix security vulnerabilities
fix_vulnerabilities.bat

2. Run the Application

streamlit run app.py

3. Access the Platform

Open your browser and navigate to:

http://localhost:8501

🔧 Development Workflow

User Workflow

Upload datasets (CSV/Excel)
Run automated profiling
Execute data quality checks
Perform reconciliation
Review AI-generated insights
Export audit report

Example Scenario

Source Dataset: 10,000 rows
Target Dataset: 9,850 rows

Findings:
- 150 missing records detected
- 138 null values identified
- Column mapping mismatch found
- AI-generated explanation provided

👥 Team Structure

Member 1: Backend Development

Data processing logic
Quality rule engine
Reconciliation algorithms

Member 2: UI/UX Design

Streamlit interface
Presentation materials
Documentation

🤖 IBM Bob Integration

IBM Bob is used as an AI development assistant to:

Accelerate coding and testing
Debug issues efficiently
Generate documentation
Provide code suggestions
Enable rapid prototyping

📊 Success Metrics

⏱️ Time saved in audit tasks
🔍 Number of issues detected
📉 Reduction in manual effort
👥 User adoption rate
✅ Accuracy improvements

🔮 Future Enhancements

Database integration (PostgreSQL, MySQL)
Role-based access control
Real-time monitoring dashboard
Advanced anomaly detection (ML models)
Enterprise audit workflow integration
Multi-user collaboration
API for external integrations

📚 Documentation

Design Document: AI_Audit_Platform_Design_Document_Full.pdf
Library Reference: LIBRARIES_SUMMARY.md
Security Report: SECURITY_VULNERABILITY_REPORT.md

External Resources

🛡️ Security Best Practices

Regular Scans: Run pip-audit weekly
Dependency Updates: Keep packages up-to-date
Virtual Environments: Use isolated environments for production
Access Control: Implement proper authentication
Data Encryption: Encrypt sensitive audit data
Audit Logging: Track all data access and modifications

🤝 Contributing

Fork the repository
Create a feature branch (git checkout -b feature/AmazingFeature)
Commit your changes (git commit -m 'Add some AmazingFeature')
Push to the branch (git push origin feature/AmazingFeature)
Open a Pull Request

📄 License

This project is licensed under the terms specified in the LICENSE file.

🏆 IBM Bob Dev Hackathon

This project demonstrates how AI and Python can transform audit workflows by improving:

✅ Efficiency: Automated data validation and profiling
✅ Accuracy: Consistent quality checks and reconciliation
✅ Scalability: Handle large datasets efficiently
✅ Transparency: AI-powered explanations and documentation
✅ Usability: Simple, intuitive web interface

📞 Support

For questions, issues, or contributions:

Open an issue on GitHub
Contact the development team
Review the documentation

⚠️ Important Notes

Security First: Always run fix_vulnerabilities.bat after initial installation
Testing: Test all updates in a staging environment before production
Backups: Maintain regular backups of audit data
Compliance: Ensure compliance with relevant audit standards (SOC 2, ISO 27001, etc.)

Last Updated: 2026-05-02 20:43 UTC Version: 1.0.0 Python Version: 3.11.3 Security Status: ✅ SECURE - All vulnerabilities resolved (0 known vulnerabilities) Status: Ready for Development

Name		Name	Last commit message	Last commit date
.streamlit		.streamlit
.vscode		.vscode
assets		assets
backend		backend
components		components
data		data
demo		demo
pages		pages
security-docs		security-docs
tests		tests
utils		utils
AI_Audit_Platform_Design_Document_Full.pdf		AI_Audit_Platform_Design_Document_Full.pdf
BACKEND_API_DOCUMENTATION.md		BACKEND_API_DOCUMENTATION.md
BACKEND_README.md		BACKEND_README.md
COLLABORATION_FEATURES_GUIDE.md		COLLABORATION_FEATURES_GUIDE.md
COORDINATOR_WORKFLOW_GUIDE.md		COORDINATOR_WORKFLOW_GUIDE.md
INTEGRATION_REPORT.md		INTEGRATION_REPORT.md
LIBRARIES_SUMMARY.md		LIBRARIES_SUMMARY.md
LICENSE		LICENSE
MERGE_PLAN.md		MERGE_PLAN.md
QUICK_START.md		QUICK_START.md
README.md		README.md
README_UI.md		README_UI.md
SAMPLE_DATA_GUIDE.md		SAMPLE_DATA_GUIDE.md
SUBMISSION_CHECKLIST.md		SUBMISSION_CHECKLIST.md
TEST_RESULTS.md		TEST_RESULTS.md
app.py		app.py
requirements.txt		requirements.txt

Latest commit

History

Folders and files

Repository files navigation