While others move AI to the cloud, we moved the Audit to the edge. JustAirAudit is the first tool designed to interpret PCI v4.0.1 standards entirely within your perimeter.
STEP 1: LOCAL REDACTION - A Python-based pre-processor scans and redacts high-entropy strings (Keys/Tokens) before the data reaches the inference engine.
STEP 2: AIR-GAPPED INFERENCE - The sanitized data is analyzed by a local Llama 3 instance. Zero-Egress: No data ever touches the cloud. No API calls. 100% On-Prem.
Process raw logs within your hardened perimeter. An integrated Python Pre-Processor automatically scrubs high-entropy strings and credentials before analysis. Zero-Egress Architecture ensures no sensitive data ever leaves. ✖️🌐
The engine utilizes a quantized Llama 3 model to parse data against the latest PCI v4.0.1 criteria. 🛡️
Generates detailed gap analysis reports without ever sending a single byte to the cloud. 🗝️
1. Download Ollama Runtime.
2. Pull Llama 3 via CMD: ollama pull llama3
3. Ensure Ollama is running in the background before execution. 🚀
4d6c38568174e2a8414da8f65c697e4f0770f2b881e6c522ffa2cdc2262c61c4
# AWS S3 Configuration - config.py
AWS_ACCESS_KEY_ID = "AKIAEXAMPLE12345678"
AWS_SECRET_ACCESS_KEY = "wJalrXUtnFEMI/K7MDENG/bPxRfiCYEXAMPLEKEY"
Analysis Input (Sanitized):
Vulnerabilities Identified:
1. UNSECURED ACCESS KEYS: Detected hardcoded credential patterns in `config.py`. Even when redacted, the presence of static assignment violates security best practices.
Compliance Gaps:
• REQ 3.5: Primary account credentials must be stored in a secure cryptographic environment.
*Inference performed locally on 127.0.0.1. No external egress detected.
