AWS FinOps Auditor (Lite)

A lightweight, local Python CLI tool designed to detect orphaned AWS resources and identify hidden infrastructure waste before it compounds into financial debt.

This repository contains the free core-logic preview (Lite Version).

🚀 Upgrade to Enterprise

Need full automation? The Enterprise AWS FinOps Auditor includes:

• Automated Multi-Region Scanning: Scans all active AWS regions simultaneously.
• Zero-Setup Credential Handling: Automatically assumes cross-account roles.
• Slack & Teams Integrations: Pushes daily/weekly waste reports directly to your engineering channels.
• 1-Click Remediation: Generates execution scripts to instantly delete orphaned resources.

👉 Download the Enterprise Engine Here

---

📖 Lite Version Overview

The Lite engine runs a local pandas pipeline against exported AWS billing or usage CSVs. It executes multi-vector financial cost-waste calculations to flag:

• Unattached EBS Volumes
• Idle EC2 Instances (CPU < 5%)
• Unassociated Elastic IPs
• Public-facing security anomalies

Quickstart

1. Clone the repository:

   git clone https://github.com/Ace7-coder/aws-finops-auditor.git
   cd aws-finops-auditor
   

2. Install dependencies:

   pip install -r requirements.txt
   

3. Run the auditor against your usage CSV:

   python infrastructure_auditor.py --file my_aws_usage.csv
   

Output

The engine outputs two files to your local directory:

• audit_report.json: Granular JSON output of all flagged resources.
• executive_summary.md: A high-level markdown summary of total waste identified (in USD) and critical security flags.

⚙️ Core Logic Architecture

def calculate_waste(self, df: pd.DataFrame) -> float:
    """Multi-vector financial cost-waste calculations."""
    # Identifying idle instances (CPU < 5%) and unattached volumes
    waste_mask = (df['cpu_utilization'] < 5.0) | ((df['amount'] > 0) & (df['cpu_utilization'].isna()))
    return df.loc[waste_mask, 'amount'].sum() if not df.loc[waste_mask].empty else 0.0

Contributing

Pull requests are welcome. For major changes, please open an issue first to discuss what you would like to change.

License

MIT