Skip to main content

Purple Fabric

Menu

Multidimensional-Multilayer System of Connected Knowledge

The AI-native engineering system for enterprise applications.

Enterprise knowledge is everywhere. But it lives in silos — disconnected, incomplete, always changing. AI without connected knowledge is intelligent, but never accurate.

MSOCK makes enterprise knowledge computationally understandable, so AI can act safely, intelligently, and autonomously — on your estate, not a guess at it.

Multidimensional Context

Multidimensional Context

Every object read against the business, operational, and technical layers it belongs to.

Accurate Relationships

Accurate Relationships

Connections backed by evidence, not inferred from proximity.

Living Knowledge

Living Knowledge

The model updates as the estate changes, so it doesn't go stale the day after capture.

Book a Briefing

For highly regulated financial institutions

MSOCK is not a feature. It's a new enterprise knowledge infrastructure essential for an AI-governed future.

Launching MSOCK

The new Enterprise Knowledge Infrastructure for Financial Services Essential for an AI-governed future

RDBMS made data Structured and Queryable.

MSOCK makes enterprise knowledge Connected and Computable

Keep your agents. Keep your AIDLC. Add MSOCK.

Most financial institutions are already building their own agents and AI development lifecycles. MSOCK coexists with that architecture and adds the understanding, accuracy, and control your AI needs to act on your enterprise instead of around it.

Your Existing Investments

AI Agents

Copilots

AIDLC

LLMs

Engineering Tools

plus

MSOCK

Understanding, accuracy, and control— laid over what you've already built.

The 5-Stage MSOCK Cognitive Engine

A progressive intelligence chain that turns enterprise knowledge into governed action.

1

Understand
SOCK

Connect and organise enterprise knowledge and relationships.

2

Visualise
SPATIAL GRAPH

See your enterprise across business, operations, and technology.

3

Assess
BLAST RADIUS

Know what a change will affect before you make it.

4

Act
ENGINEERING

Build, modernise, change, and run — with enterprise context.

5

Govern
COMMAND CENTRE

Oversight, approvals, risk, audit, and outcomes on autonomous action.

1. UNDERSTAND · SOCK, SYSTEM OF CONNECTED KNOWLEDGE

Know how your enterprise system actually works.

The truth about your software is scattered across code, schemas, configuration, documents and people's memory. SOCK reads that estate and reconciles it into one connected view, with the evidence behind every link kept in place.

Complete

The whole ingested scope, not one repository at a time.

Connected

The relationships between things,
not a list of them.

Evidenced

Every claim opens to a file
and a line.

Current

Refreshes after releases, so it follows the software.

What one capture holds, layer by layer

L1

User journeys and UI

Captured Journeys, UI docs, Screen Flows, Click Paths

L2

Service interactions

Call chains, Multi-hop API contracts, Auth, Failures

L3

API contracts

API contract docs, Endpoints, REST + GraphQL + gRPC

L4

Service Architecture

Services, Arch docs, Tech stack, Deployment Config

L5

Data Models & Databases

Database schemas, Tables, Collections, Constraints

L6

Operations

Playbooks, Environments, Observability, Error Catalogs

L7

Infrastructure & Deployment

Infra docs, Distribution docs, CI/CD, k8s Helm charts

Seven layers, one capture, one connected view — read from the user's screen down to the operational runbook.

2. VISUALISE · 21 DIMENSION SPATIAL GRAPH

The system as it actually runs — not as anyone assumes it does

A requirement enters at the top of the estate and its impact travels all the way to the bedrock.
That is why what does this change touch and what breaks if it fails have never had an honest answer — most tools only see what sits above the waterline.

21 dimensions of enterprise ground truth

D1-D3

Experience & Intent

  • Business journeys.
  • UI flows
  • Capability maps

D4-D7

Governance & Trust

  • Business rules.
  • Compliance footprints
  • Security posture
  • Reliability maps

D8-D12

Execution & Connectivity

  • Service interactions
  • Call chains
  • API contracts
  • Data relations
  • Lineage

D13-D17

Infrastructure & Estate

  • Infra inventory
  • Operations playbooks
  • Integration topologies
  • Estate maps
  • Dependancy blast radius

D18-D21

Archaeology & Seams

  • As-is archaeology
  • Migration seams
  • Interface mappings
  • Dead code detection

What it costs you today

Product, architecture, security, compliance, data and operations each work from a different picture of the same estate. Context is lost at every hand-off, and a change that looks local at the journey level can cross controls, interfaces and legacy seams underneath it — every local decision looks reasonable, and they still collide at a shared boundary.

What the Spatial Graph does

It gives each person a view built for the question they're asking, over one connected model. Start from a journey, rule, capability, service, data object, alert or migration seam. Add only the dimensions that matter, then follow any relationship down to the evidence behind it — the source, not a summary of it.

What this is not

The point isn't twenty-one dimensions on one screen, and 21D isn't a twenty-one-axis picture. Product and engineering stop receiving separate diagrams that must be reconciled later — they're looking at different views of the same objects. The graph is navigation; the evidence sits behind it.

3. ASSESS · PRECISE BLAST RADIUS

Turning change uncertainty into
deterministic certainty

Every change to a live system raises one question: what else does this touch? Most tools answer by tracing every technical connection they can find — treating “connected” as the same thing as “affected.” That habit quietly costs enterprises twice over: real risk gets missed, or every change starts to look enterprise-wide and teams end up testing everything “just in case.”

MSOCK calculates the actual blast radius of a change before a single line of code is written — grounded in how the system truly connects, across customer journeys, processes, policies, services, APIs, events, data, code, and tests.

Assess Precise Blast Radius

Why the answer can be trusted

The output isn’t a red circle drawn around a diagram. Every item comes with the relationship chain that explains why it’s included, the evidence behind it, and a confidence level — so a reviewer can inspect, question, or rule out any result before it drives a decision.
The output isn’t a red circle drawn around a diagram. Every item comes with the relationship chain that explains why it’s included, the evidence behind it, and a confidence level — so a reviewer can inspect, question, or rule out any result before it drives a decision.
  • The change
  • Direct hits — must change
  • Regression scope — needs retesting
  • Safe zones — left untouched

Where this shows up

  • A new regulation, or a change to an existing rule: see exactly which journeys, code, data, and controls it touches — across privacy, tax, reporting, eligibility, or retention.

  • Retiring or versioning an API, with confidence about who actually depends on it, instead of a best guess from a spreadsheet.

  • A recurring incident, traced to the exact services, releases, configurations, and journeys it could be implicated in.

  • A product launch, checked against an in-flight platform or database migration — before the two collide in production.

  • A move to the cloud, with every downstream file, interface, control, and operational procedure it could disturb mapped out in advance.

    • Zero regression risk. No blind testing. No guessing at scope. Just confidence before code is touched.

    4. ACT · FOUR ENGINEERING MACHINES

    Four ways to build, fix, run and modernise your enterprise system

    Most AI coding tools treat every task the same way: a prompt in, code out. But building something new, changing something live, fixing an incident, and retiring old technology are different jobs — each needs different expertise, method, and proof. MSOCK applies one connected, evidence-backed understanding of your systems to four purpose-built engineering machines, so no task starts from zero.

    Build

    NEW APPLICATIONS

    Launch new digital products without reinventing the wheel — reusing the identity, payments, fraud, and data capabilities you already have.

    For example — new applications, new modules, new capabilities and features.

    Change

    CHANGE REQUESTS

    See exactly what a regulatory update, pricing change, or feature request will touch — across every affected system — before a single line of code is written.

    For example — rolling out a new compliance rule across several applications, with tests and evidence ready to prove it was done safely.

    Run

    FASTER BUG FIXES

    Connect an alert straight to the logs, the recent deployment, the code owner, and a likely fix — cutting investigation time when something breaks.

    For example — tracing a recurring payment error back to the exact release and line of code that caused it.

    Modernise

    LEGACY, WITHOUT THE GUESSWORK

    Separate the business logic trapped in an old system from the technology it runs on, then rebuild it on modern architecture — with proof nothing changed for the business.

    For example — moving a mainframe workload to the cloud one verified piece at a time, with evidence behaviour hasn't shifted.

    The honest boundary

    RUN supports the people running your systems — it doesn’t process live transactions. Modernise isn’t a one-shot rewrite — it’s a verified, step-by-step migration.

    Probabilistic intelligence, inside deterministic control

    Shared health checks, compliance guardrails, and automatic circuit breakers apply across all four machines. MSOCK is part of the Purple Fabric platform, by
    Intellect Design Arena.

    5. GOVERN · COMMAND CENTRE

    Make the development cycle scalable, repeatable and governed

    The method your organisation already follows — carried with the work, instead of left behind in a document.

    The Problem

    Engineering decisions, AI actions, evidence, and approvals scatter across tickets, repositories, pipelines, chats, and meetings. A work tracker can tell you that a task moved. It can't tell you what had to be true for that move to be allowed — and as AI does more of the work, that gap only widens.

    The Command Centre

    It orchestrates people, process, disciplines, knowledge and tools through a governed, auditable lifecycle, making AI-native engineering repeatable and scalable.

    Personas on the lifecycle

    Product Owner

    Guild

    Capability Lead

    Engineer

    Quality Assurance

    Release Manager

    DevOps

    Security

    Five phases, Four gates

    Ideation

    Product Owner : Guild

    PRD + HLD

    AI Generates

    G1

    Inception

    Capability Lead : Guild

    PRD + Stories + LLDs

    AI Decomposes

    G2

    Construction

    Engineer : Security

    Code + Tests

    AI Implements

    G3

    Validation

    Engineer : QA

    Deploy + Certify

    AI Pre-flight

    G4

    Operations

    Release Manager : DevOps

    Release + Learn

    AI Captures Learning

    What this is not

    Another status dashboard. It doesn't ask you to replace your trackers, repositories, or pipelines — it carries the method across the tools you already use.

    Grounded in evidence • Guided by intelligence • Governed by design.