Architecture preview
The security operating system for AI
Reasoning connects to action. Evidence shapes what the system remembers. Guardrails bound what it can do.
A proposed composition of models, agents and controls.
Inspect the architecture- ReasoningSecurity-focused multimodal reasoning
- Agent fabricPlanner / hunter / responder / critic
- Experience compilerScoped, revocable skills and tools
- GuardrailsContinuous testing before and during action
The system design
Authority stays explicit.
Inspect a layer to see the boundary it adds. The shared fabric carries context and provenance across the design.
Reasoning
Freedom v1.1
The model sits inside the proposed system as its reasoning layer. Model output is input to the governed agent workflow; it does not carry execution authority by itself.
- Image + text input
- Autoregressive text output
Hover a layer, or focus it with Tab, to inspect its role.
- Freedom v1.1
- Agent fabric
- Experience compiler
- Guardrails
Shared data fabric
Normalisation, ontology, risk graph and provenance ledger
Component vocabulary in the proposed design.
- Swarm Fabric
- Conductor
- ProveIt
- Fabric
- Ontos
- RiskGraph
- AgentMesh
- Provenance Ledger
Proposed assurance loop
Our agents tested our own security products
Agents exercise a bounded target. Independent verification evaluates the result. Supported experience may be compiled; contradictory outcomes can revoke it. This diagram describes a design, not completed test results.
Provenance and independent evidence govern reuse.
Domain coverage · D01–D34
A map of the ambition.
Explore the AI security site. The remaining domains describe roadmap scope, with no product availability implied.
- D01
AI & Agentic Security
Explore
- D02
AI for Security
Roadmap
- D03
Offensive Security & Adversary Emulation
Roadmap
- D04
Exposure & Vulnerability Management
Roadmap
- D05
Application & Product Security
Roadmap
- D06
DevOps, Platform Engineering & Internal Developer Platform Security
Roadmap
- D07
Client-Side, Browser-Extension & Frontend Supply-Chain Security
Roadmap
- D08
API Economy & B2B Integration / SaaS-Mesh Runtime Security
Roadmap
- D09
Hardware, Firmware & Supply-Chain Security
Roadmap
- D10
Cloud & Cloud-Native Security
Roadmap
- D11
Network Security
Roadmap
- D12
Availability, DDoS Protection & Edge Resilience
Roadmap
- D13
Endpoint, Device & Mobile Security
Roadmap
- D14
Observability-Security Convergence
Roadmap
- D15
Identity & Access Security
Roadmap
- D16
Data Security & Privacy Engineering
Roadmap
- D17
Privacy-Enhancing Technologies (PETs) Platform
Roadmap
- D18
Detection, Response & SecOps
Roadmap
- D19
Cyber Resilience, Recovery & Ransomware Continuity
Roadmap
- D20
Threat Intelligence, Fraud & Digital Risk
Roadmap
- D21
Trust & Safety, Content Integrity & Platform Abuse
Roadmap
- D22
Email, Collaboration & Human Comms Security
Roadmap
- D23
Cryptography, PKI & Post-Quantum
Roadmap
- D24
Cryptographic Incident Response & CRQC Readiness Operations
Roadmap
- D25
GRC, Risk & Compliance
Roadmap
- D26
Financial Crime, AML, KYC & Transaction Monitoring
Roadmap
- D27
Human Layer, Insider Risk & Education
Roadmap
- D28
OT / ICS / IoT / Cyber-Physical
Roadmap
- D29
Physical Security & Cyber-Physical Convergence
Roadmap
- D30
Telecom, Carrier & Signaling Security
Roadmap
- D31
Web3 / Blockchain / Smart Contract Security
Roadmap
- D32
National & Critical Cyber Defense
Roadmap
- D33
Frontier & Emerging Threats
Roadmap
- D34
The Security OS — Unifying Platform & Data Fabric
Roadmap
Build with clear boundaries
Start with your system.
Discuss the models, agents and authority boundaries in your environment.