PLATFORM ARCHITECTURE

The AI Agent Security Graph & Control Plane

Gaussian unifies AI client discovery, MCP server mediation, skill package auditing, and kernel execution authority into a single control plane across developer workstations.

Core Platform Architecture Modules

Discovery

Continuous Asset Discovery

Automated scanning of mcp.json configs, SKILL.md packages, and 35+ AI client process signatures.

View Technical Guide

MCP

MCP Security & Governance

Discovery, risk tiering, and tools/call policy mediation inline with secret redaction.

View Technical Guide

Skills

Agent Skill Governance

Audit SKILL.md manifests, prompt injection safety, and embedded scripts/*.py helpers.

View Technical Guide

macOS

macOS Sensor & Exec Firewall

Endpoint Security (ES) and Network Extension (NE) protection for terminal commands & egress.

View Technical Guide

AI AGENT SECURITY GRAPH · LIVE RUNTIME MAP

Interactive Capability & Execution Topology

Gaussian_ Execution graph · macOS

Deterministic Governance Observatory

Constraining autonomous execution through layered deterministic governance. Watch real-time agent intents being evaluated against hard policy boundaries.

Threat Classification

Low (Read/Describe)

Medium (Config/Scale)

Critical (Mutate/Delete/Secrets)

Enforcement Protocol

CONTAINMENT ACTIVE

Live Intercept Stream

Platform Expansion Roadmap

Shipped

Phase 1

macOS

Endpoint Security (ES) + Network Extension (NE) system extensions. Native Apple APIs, no KEXT.

2027 Roadmap

Phase 2

Windows

ETW (Event Tracing for Windows) + WFP (Windows Filtering Platform) for enterprise Windows developer fleets.

2027 Roadmap

Phase 3

Linux / CI

eBPF-based process tracing for Linux developer workstations and CI/CD pipeline runners.

Future

Phase 4

Cloud Runtime

Agent governance for cloud-hosted AI pipelines, SaaS tool orchestrators, and remote execution environments.