SEE IT IN ACTION
The Challenge
The senior architect who built the repository knows why every convention exists. When they leave, the conventions leave with them. Junior architects work from what they absorbed informally. Drift begins within months, and there is no mechanism to catch it at creation time.
Word documents and slide decks describe standards. They do not enforce them. Every model Sparx EA accepts regardless of what the document says. Governance review catches violations after the model is built, not while it is being built. The feedback loop is days.
Two days of facilitated agreement produces a list of conventions. The list goes into a document. The document does not integrate with the modeling tool. Six months later, the team is working from memory again.
When you use AI Augmented Architecture for Reference Architecture
AI Power Tools for EA runs schema analysis across your live repository: stereotypes in use, connector patterns, tagged value usage, base metaclass patterns. The output is a structured picture of your team's actual operating conventions, organized for adjudication. No recall required. No workshop to schedule before the preparation work is done.
The observed patterns are candidates. The architect classifies each one: intentional standard, legacy drift, or needs a new alias. Claude records the adjudication and builds the intermediate metamodel. Every standard decision belongs to your architect. The tool observes and presents; the architect decides.
Claude composes the MDG XML and installs it in your repository. The repository now has a formally defined MDG built from what your team was already doing. Not authored from scratch. Not borrowed from a generic template. Built from your repository.
After install, AI Power Tools for EA runs a conformance pass. You see exactly how many existing elements conform to the new standard and what the outlier categories are. You have a scoped cleanup plan before you commit to the standard.
The MDG is installed in Sparx EA. It is not a document someone needs to remember to consult. When your senior architect leaves, the conventions they built stay in the tool, enforced at creation time, for every architect on the team.
AI POWER TOOLS FOR ENTERPRISE ARCHITECT
86 tools. Full read/write access to your Sparx EA model. Seven-day free trial. No credit card required to start.
HOW TO GET STARTED
Understand your starting point
Get AI Power Tools for EA running
Build the habit across your team
FREQUENTLY ASKED QUESTIONS
The MDG is built from your repository's observed patterns, adjudicated by your principal architect. It reflects the conventions your team has been applying, confirmed as intentional by the people who built them. It is not a generic Sparx EA best-practice template.
Drift and inconsistency are exactly what the schema analysis surfaces. Patterns with low usage counts or inconsistent application are presented to the architect alongside well-established conventions. The architect excludes drift patterns from the MDG during adjudication. The conformance baseline after install shows the scope of cleanup for the existing models.
The adjudication session is significantly better when your senior architects participate. They can classify patterns from knowledge rather than inference. If a senior architect is leaving, running the engagement while they are still available is the recommended approach. The toolchain can still run the observation analysis after they leave, but the adjudication session will require more research.
A valid Sparx EA MDG XML file, named and versioned by your organization. It is installed in your Sparx EA environment and can be accessed from all architects' workstations. It is yours to maintain, extend, and version going forward.
Sparx EA's Profile tooling lets you author an MDG manually, stereotype by stereotype. AI Power Tools for EA automates the observation step (reading the patterns from the existing repository) and the composition step (writing the XML from the adjudicated set). The result is the same kind of MDG file that Sparx EA's Profile tooling would produce, but built from observed evidence rather than from authoring from scratch.
They are in the repository. AI Power Tools for EA reads them, your principal architect confirms which ones are standards, and the result is an MDG installed in Sparx EA before the people who built the conventions move on.
Schedule a Discovery Call