Sentinel: Industrial Edge Systems & Process Governance

Tech Stack: Python 3 | AutoHotkey v2 | Windows API (ntdll, kernel32) | REST APIs | ETL Pipelines

Executive Summary

Sentinel is a monorepo containing a suite of high-assurance automation tools, IoT telemetry pipelines, and kernel-level process governors. It was engineered to bridge the gap between Operational Technology (OT) and Information Technology (IT) on high-volume Surface-Mount Technology (SMT) manufacturing lines.

In legacy manufacturing environments, systems are often black boxes, APIs are undocumented, and race conditions between proprietary binaries can bring production to a halt. The tools in this repository were built to establish strict digital governance over these physical systems—reducing machine downtime, neutralizing configuration drift, and automating heavy data ingestion pipelines.

Repository Architecture

This repository is divided into four distinct domain-specific modules:

1. Warden - Kernel-Level Process Governance

A high-assurance data sanitization proxy built to combat severe configuration drift in SMT inspection hardware.

• The Engineering: Utilizes ntdll\NtSuspendProcess to hijack and freeze proprietary C-binaries at the CPU level. This "nuclear" synchronization resolves non-deterministic race conditions, allowing Warden to safely sanitize, normalize, and inject overridden states into I/O files before releasing the process back to the system.
• The Impact: Prevented thousands of false-positive machine failures, saving the factory floor from years of manual configuration debt.

2. Nozzle Suite - IoT Telemetry & Multithreaded Dashboards

A complete IoT suite that reverse-engineers undocumented hardware APIs to provide real-time tooling tracking and ETL inventory reporting.

• The Engineering: Consumes deeply nested, undocumented JSON payloads from Fuji NXT machine REST APIs. Translates internal hexadecimal machine codes into human-readable inventory SKUs. Serves this data asynchronously to a multithreaded tkinter desktop GUI, completely decoupling heavy network I/O from the user interface.
• The Impact: Replaced manual tooling searches, significantly reducing Single-Minute Exchange of Die (SMED) downtime during PCB changeovers. Now serves as the mandatory Standard Operating Procedure (SOP) for all operators.

3. Orchestrator - Temporal Data Synthesis Pipeline

An automation engine designed to rapidly clone, offset, and inject inspection metadata into highly rigid legacy SQL databases.

• The Engineering: Parses CSV metadata via RegEx and manipulates temporal data using custom base-60 math functions. Implements strict Buffer Overflow Protection/Temporal Normalization by mathematically handling 24-hour epoch rollovers, ensuring simulated timestamps never crash the database ingestion engine.
• The Impact: Turns hours of manual data entry ("carpetazo") into a sub-second, error-free automated process.

4. KeepAwake - Session Persistence Daemon

A lightweight daemon designed to bypass aggressive global group policies (GPO) on factory floor terminals.

• The Engineering: Directly interacts with the Windows API via kernel32\SetThreadExecutionState (0x00000002). By programmatically asserting the ES_DISPLAY_REQUIRED flag, it mimics human presence without relying on simulated keystrokes or mouse wiggles.
• The Impact: Prevents critical monitoring dashboards from locking out operators during live production runs.

Core Engineering Philosophies

• Pragmatic Problem Solving: Solutions are dictated by physical factory constraints (e.g., frozen UIs, physical tool searches, database crashes), not academic theory.
• System Observability: Creating transparency in undocumented or legacy systems where none previously existed.
• Defensive Architecture: Using temporal offsets, kernel suspension, and strict network timeouts to ensure high-assurance stability in chaotic environments.

Getting Started

Each module is self-contained. To explore the code or deploy a specific tool, navigate to its respective directory and review its localized README.md for specific configuration, dependencies, and execution instructions.

Disclaimer

Please note that this repository functions as an architectural outline. While the logic flows and system interactions (e.g., ntdll hooks, IoT ETL pipelines) are functionally accurate to the actual Sentinel framework, the production-grade binaries used in industrial settings are distinct and heavily customized.

Users are advised that direct deployment of these scripts is not recommended. The author assumes no legal responsibility for system instability, hardware damage, or data loss resulting from the use of this reference code.