AutomatosX

Name: AutomatosX
Author: defai-digital

Sovereign AI orchestration & governance platform

AutomatosX sits above inference engines: application entry points, governance, and private AI infrastructure — for enterprise, government, and defence-grade operations.

This repository introduces the AutomatosX system. Product source and releases live in dedicated repositories.

System overview

Open the interactive one-pager (architecture, products, data flow):

→ assets/AutomatosX_Infographic.html

AutomatosX system infographic


Interactive	AutomatosX_Infographic.html — open in a browser
Static	AutomatosX_Infographic.png
Positioning	Contract-first · deterministic · offline / air-gapped ready
Role of this repo	System introduction and product map

Products

Public (open source)

Product	Role	Repository
AX Studio	General GenAI workspace — enterprise knowledge + agentic workflows	defai-digital/ax-studio
AX Code	AI coding & automation — developer / IT agent runtime	defai-digital/ax-code
AX Engine	Mac-native inference — Apple Silicon optimized local models	defai-digital/ax-engine

Private (enterprise)

Part of the same system architecture — no public source links.

Product	Role
AX Trust	Deterministic AI pipeline — contracts, guardrails, audit, explainability
AX Serving	Enterprise orchestration — multi-node routing, heterogeneous compute
AX Fabric	Knowledge infrastructure — RAG, distillation, knowledge lifecycle

Architecture, layers, and system data flow are in the infographic above.

Why AutomatosX

Pillar	Meaning
01 · Verifiable AI	Decisions are traceable, inspectable, and anchorable
02 · Reproducible pipelines	Contract-based execution — same input → same output
03 · Full audit trail	Explainability across reasoning steps and data flows
04 · Sovereign / offline	Air-gapped capable — data sovereignty for sensitive ops

Governance above LLMs — not only inference (vLLM, llama.cpp, …); AutomatosX adds governance, orchestration, and reproducibility
Sovereign / offline AI — air-gapped paths; Mac + edge (e.g. Jetson Thor) hybrid designs
Deterministic vs only probabilistic — contract-first execution for reliability and audit
Heterogeneous compute — Mac Studio · Jetson Thor grid · extensible to DGX-class clusters

Documentation

Doc	Contents
docs/introduction.md	What AutomatosX is, who it is for
docs/system.md	Full system architecture (layers + data flow)
docs/products.md	Product directory (public + private)
docs/getting-started.md	Start with Studio, Code, or Engine
docs/glossary.md	Terms

Quick start

AX Studio — GenAI workspace

https://github.com/defai-digital/ax-studio

Local-first desktop workspace: chat, providers, MCP tools, memory, optional local inference via AX Engine.

AX Code — coding agents

https://github.com/defai-digital/ax-code

Agent runtime for real repositories: TUI, CLI, SDK, sandboxing, sessions, MCP.

AX Engine — local models on Apple Silicon

https://github.com/defai-digital/ax-engine

Install, prepare models, serve OpenAI-compatible endpoints on Mac — then point Studio or Code at Engine.

Organization

Built by DEFAI Digital · github.com/defai-digital


Product	automatosx.com
Company	defai.digital
System docs	This repository

Repository layout

automatosx/
├── README.md                              # System introduction
├── assets/
│   ├── AutomatosX_Infographic.html        # Interactive system one-pager
│   └── AutomatosX_Infographic.png         # Static preview for GitHub
├── docs/                                  # Architecture & guides
└── AGENTS.md

License

Documentation here is for system orientation. Each product repository carries its own license — check the product repo before installing or redistributing.