Skip to content

Security Overview

Nyroxis is built with a strict security-first philosophy: all operations are local, encrypted, tamper-resistant, and designed to protect the user without exposing any data to third parties.

This overview summarizes the main security principles and mechanisms used across the platform.


Core Security Principles

1. Privacy by Design

Nyroxis does not upload, sync, or transmit any data to servers. All security logic, monitoring, detection, and AI analysis run entirely on the device.

2. Encryption Everywhere

Every captured event is: - Encrypted instantly at capture (AES-256) - Stored only in encrypted form — never in plaintext - Decrypted exclusively in memory during processing - Protected with integrity hashes

No plaintext logs ever exist on disk.

3. Zero Cloud Dependency

Nyroxis does not rely on: - Cloud storage or processing - External APIs - Remote servers - Telemetry pipelines - Online authentication

Users maintain full autonomy and zero external exposure.

4. Multi-Layer Detection

Nyroxis Intelligence provides: - 27 detection rules — known threat patterns - 12 correlation rules — multi-event patterns over time - 2 chain rules — multi-stage attack sequences - Extensible by security professionals without modifying core components

5. Local AI/ML

The AI/ML engine processes data locally: - Isolation Forest anomaly detection - Z-Score statistical classification - Contributing feature identification

No cloud training, no cloud inference, no data sharing.


Key Security Mechanisms

1. Secure Event Capture

Nyroxis Agent captures: - Processes and services - File system and registry changes - Network connections - Privilege actions - Windows Event Log entries

All events are encrypted immediately at capture.

2. Tamper-Resistant Storage

The local SQLite database includes: - AES-256 encryption - Hash-chained event blocks - Protected write paths - Integrity verification on every read

Attackers cannot modify, delete, or inject logs without detection.

3. Platform Resilience

Nyroxis System Guardian: - Monitors Agent and Intelligence every 3 seconds - Detects unexpected service stops immediately - Logs shutdown attempts as security events - Manages backups and offline license validation

The platform cannot be silently disabled.

4. HWID-Based Licensing

License validation: - Bound to device hardware (HWID-derived key) - Validated offline using AES-GCM and HMAC - Enforced by System Guardian — services stop automatically if license is invalid

5. Dashboard Security

The Dashboard: - Reads decrypted summaries only in RAM - Never writes back decrypted data to disk - Does not connect to external sources - Provides read-only access to log storage


Security Layers Summary

Nyroxis integrates multiple defensive layers: - AES-256 encryption at capture - Hash-chained tamper-resistant storage - Platform guardian for service resilience - Local-only AI/ML engine - Offline license validation - Zero cloud exposure - Privacy-first architecture throughout


Summary

Nyroxis provides a modern, private, offline security model that protects users without tracking them — combining transparency, control, forensic integrity, and strong defensive design at every layer.

© Nyroxis Documentation — Nyroxis® Endpoint Intelligence Platform
Built with MkDocs Material — Documentation auto-generated from public modules.