Kernaro AI Hub vs Prolaborate: Which Stakeholder Layer Is Right for Your EA Practice?
Put plainly: these two products answer different questions, so lining them up head to head is the wrong frame. Prolaborate is a publishing and browsing layer — it takes content from your EA repository and presents it to non-architect stakeholders in a structured, navigable portal. Kernaro AI Hub is an AI intelligence layer — it lets those same stakeholders ask the repository natural-language questions and get governed, contextual answers back. Prolaborate users browse what architects have published; Kernaro users ask what they want to know. For many organizations the honest answer is both. Where budget forces a single choice, decide on one thing: do your stakeholders mostly need to browse, or mostly need to ask?
Both products exist because the same problem keeps biting EA teams — the people who most need architecture answers are the ones least able to open Sparx EA. The two solve it from opposite ends. Below is how they actually differ, and where each earns its keep.
The core distinction
Strip away the feature lists and the difference is a single contrast: a portal you read versus a conversation you query.
Prolaborate — publish and browse
A web portal that takes what your architects have curated and renders it for stakeholders. The architect decides what is visible and when; the stakeholder navigates it.
- Architecture diagrams as rendered, interactive views
- Element detail — names, descriptions, tagged values — in a structured layout
- Navigation by package, by diagram, by search
- Filtered views scoped to a particular stakeholder group
- Report-style outputs drawn from the repository
User journey: the architect publishes from Sparx EA; the stakeholder opens a browser and reads a pre-structured presentation layer.
Kernaro AI Hub — ask and answer
A conversational interface that connects an AI assistant to the live repository. Instead of browsing published pages, the stakeholder types a question and gets a synthesized answer.
- “Which applications support the Customer Onboarding capability?”
- “What is the current lifecycle status of our CRM applications?”
- “Which technology components have no documented owner?”
- “What decisions came out of the last Architecture Review Board?”
- “Which capabilities have no application coverage?”
User journey: the stakeholder asks; the assistant queries the repository over MCP, retrieves governed data, and answers. No browsing, no pre-published views required.
What Prolaborate is
Prolaborate is a stakeholder engagement platform built and sold by Sparx Systems. It connects to a Sparx EA repository and publishes architecture content — diagrams, element lists, models, reports — to a browser portal that people can use without a Sparx EA license.
It tends to land with business stakeholders who need to see deliverables but have no EA license, executive sponsors reviewing published views, project managers checking application-portfolio data, and operations teams browsing the technology landscape. The common thread is consumption: someone curated a view, and the stakeholder reads it.
On licensing, Prolaborate is purchased directly from Sparx Systems — it is a separate product from Sparx EA and is not a Sparx Services offering. Sparx Services can advise on configuration and integration as part of a wider enterprise-architecture program, but the commercial relationship is between your organization and Sparx Systems.
What Kernaro AI Hub is
Kernaro AI Hub is a web-based portal that gives business stakeholders guided, AI-assisted access to EA content without a Sparx EA license. Under the hood it uses the MCP interface of EA GraphLink to pull live model data, then presents it through a conversational interface. Rather than browsing pre-published pages, stakeholders ask questions and the assistant answers from the repository as it stands today.
It suits executives who want architecture intelligence on demand without navigating a portal, project managers probing system dependencies, architects spot-checking their own repository for governance gaps, and analysts tracing which systems support which business processes.
Kernaro AI Hub is a Sparx Systems product, and it is new — it launched in January 2026, so any team running it today is measured in its first weeks and months, not years of production use. It does not stand alone: deploying it depends on EA GraphLink and its MCP server interface being configured first. Sparx Services helps you stand up that EA GraphLink foundation and get the repository to the standard the AI layer needs; if you want to connect AI directly to the model your architects work in day to day, that is the role of the local AI Power Tools for EA MCP server, which is a separate product.
The technical distinction
Both products read from the same Sparx EA repository. What separates them is how they read it and what they hand back.
Prolaborate uses Sparx EA's native data access to render architecture content as structured HTML views. It shows you what the repository contains, formatted for reading.
Kernaro AI Hub reads through EA GraphLink's MCP interface, which exposes the repository as a Model Context Protocol source. To do that, EA GraphLink needs an MDG Technology defined for the repository — the layer that translates the physical Sparx schema into the GraphQL schema the MCP server serves. To be clear, none of this is built into Sparx EA core — there is no native MCP server in the box. EA GraphLink is the paid product that adds it: a read-only server deployed centrally for enterprise-wide access. A connected assistant can then query elements, relationships, tagged values, and documentation, and synthesize an answer rather than display raw rows.
Repository governance matters for both, but it matters more for Kernaro AI Hub. Prolaborate can publish poorly governed data as a tidy list and the gaps stay hidden. Kernaro generates intelligence from that data, so missing owners, inconsistent lifecycle values, or incomplete capability coverage all surface in the answers. That is why MDG governance — the focus of Sparx Services' Configure the Solution track — is usually a prerequisite for a Kernaro AI Hub deployment that stakeholders can trust.
Comparing the user journeys side by side
| Aspect | Prolaborate | Kernaro AI Hub |
|---|---|---|
| Interaction model | Browse and navigate | Ask and answer |
| Content source | Published views | Live repository queries |
| Requires architect to publish? | Yes — content must be published first | No — queries the repository directly |
| Non-technical stakeholder friendly? | Yes | Yes — natural language |
| Dynamic queries | No — shows published content | Yes — any query the model supports |
| Answer synthesis | No — displays raw content | Yes — the AI synthesizes answers |
| Governance dependency | Moderate | High — MDG quality is critical |
| License source | Sparx Systems | Sparx Systems — on an EA GraphLink foundation Sparx Services configures |
| Maturity | Established | New — launched January 2026 |
When to choose Prolaborate
Reach for Prolaborate when:
- Stakeholders mainly need to browse and navigate published architecture views
- Your team produces structured deliverables — landscape diagrams, standards registers, capability maps — that stakeholders reference repeatedly
- You want controlled publishing, where architects decide what is visible and when
- You are not yet ready for AI integration; Prolaborate carries lighter governance prerequisites
- Your stakeholders are comfortable with portal-style navigation
- You are trying to break the habit of emailing static slide exports
When to choose Kernaro AI Hub
Reach for Kernaro AI Hub when:
- Stakeholders want to ask specific, open-ended questions rather than browse pre-structured content
- Your repository is well governed — MDG stereotypes applied consistently, key tagged values populated
- You want to surface EA answers inside the AI tools stakeholders already use, such as Copilot, Claude, or Gemini
- You are building AI-driven EA workflows where architecture data feeds broader automated processes
- Executives want architecture intelligence without the friction of learning a portal
When to choose both
The strongest stakeholder layer rarely picks a side. It runs Prolaborate for structured browsing — the technology-landscape portal, the standards register, the published capability map — and Kernaro AI Hub for dynamic intelligence — executive queries, governance checks, ad-hoc analysis. That pairing matches how people actually consume information: sometimes you want to study a map, sometimes you just want to ask a question. The bigger picture of how this fits an AI-enabled practice is laid out in AI Augmented Architecture.
Frequently asked questions
Can Kernaro AI Hub replace Prolaborate entirely?
Not entirely. Kernaro AI Hub is built for natural-language querying and answer synthesis; Prolaborate is built for structured navigation and browsing of published views. Stakeholders who prefer visual exploration of diagrams will favor Prolaborate; those who want direct answers will favor Kernaro. The two complement each other rather than replace one another.
Does Kernaro AI Hub require EA GraphLink?
Yes. Kernaro AI Hub runs on EA GraphLink's MCP interface. EA GraphLink has to be deployed and its MCP server enabled before Kernaro AI Hub can be configured. Sparx Services helps you stand up the EA GraphLink foundation and ready the repository so the Kernaro layer can sit on top. If your goal is to connect AI to the model your architects edit directly, that is a different path — the local AI Power Tools for EA MCP server.
Is Prolaborate sold by Sparx Services?
No. Prolaborate is a Sparx Systems product, licensed directly from Sparx Systems. Sparx Services can advise on Prolaborate configuration and integration as part of an EA program, but the license and vendor relationship sits with Sparx Systems.
What if my Sparx EA repository has poor data quality — does Kernaro AI Hub still work?
It still works, but its output quality mirrors the repository's data quality. If elements lack owners, lifecycle tagged values are unpopulated, or MDG stereotypes are applied inconsistently, the answers will be incomplete or imprecise. That is why Sparx Services treats MDG governance — addressed in Configure the Solution — as a prerequisite. The MDG quality gate is what turns raw repository data into trustworthy, AI-queryable intelligence.
Which AI assistants does Kernaro AI Hub support?
Because it uses the Model Context Protocol, any MCP-compatible assistant can connect — Claude, Microsoft Copilot, Google Gemini, Salesforce Agentforce, and others. MCP is an open standard, so Kernaro AI Hub is not tied to a single AI vendor; you use whichever assistant fits your ecosystem.
Can stakeholders use Kernaro AI Hub directly in Microsoft Teams or Outlook?
Yes, where Copilot is the connected assistant. Because Kernaro AI Hub uses MCP and Copilot is MCP-compatible, people with Copilot in M365 (Teams, Outlook, SharePoint) can ask EA questions inside those surfaces. The EA repository becomes a context source for the same Copilot they already use — no separate interface needed.
If you are starting from Prolaborate and want to add AI intelligence on top, Kernaro AI Hub layers onto your existing Sparx EA environment rather than replacing it. Talk to Sparx Services about Kernaro AI Hub →
Browse, ask, or both — which layer fits your stakeholders?
Talk to a practitioner about whether Prolaborate, Kernaro AI Hub, or both belong on your Sparx EA repository — and what your repository needs to be ready for AI querying.
Book a call →Keep reading
You might also be interested in
What Is Prolaborate? Sparx EA's Stakeholder Portal Explained
The publishing layer in depth — what it shows, who it serves, and how it is licensed.
Read → InsightEA GraphLink vs Manual Data Export: The Case for Live Architecture Data
Why a live connection beats exporting and re-keying the repository by hand.
Read → OfferingAI Power Tools for EA
The local EA MCP server that connects AI directly to the live repository your architects work in.
Explore → For leadersConfigure the Solution
Get repository governance to the standard AI querying actually needs.
See how →