Project

stig-ai-pipeline

Vendor-neutral tooling that turns DISA STIG XCCDF into structured, severity-ranked remediation (Markdown, JSON, and CSV), and can check a machine under a human approval gate. MIT licensed, free to run, no API key required.

At a glance

Windows profiles and extractors are built and unit tested. A live Windows host scan has not been run.

Published validation (August 2026)

The HackerNoon article Building a Reusable AI Skill for STIG-to-Excel Conversion records validation across three system classes totaling 522 rules (Windows Server 2019, SQL Server 2022 Instance, Apple macOS 26) with no parsing anomalies.